Compact after-loading modular heat exchangers and related methods

By adopting the design of post-installed compact air-liquid heat exchanger and heat exchanger in the computing system, the existing liquid cooling system has solved the problem of high cost and complexity in high-performance computing systems, and efficient heat exchange has been achieved and reduce the demand for environmental cooling systems.

CN119002641APending Publication Date: 2024-11-22INTEL CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311552157.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-06-29
Filing Date
2023-11-21
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Existing liquid cooling systems have high manufacturing costs and complexity in high-performance computing systems, increasing system cost and complexity, and making it difficult to effectively cool the power components of the computing system.

Method used

Using a compact air-liquid heat exchanger after installation, cooling the thermal exhaust of the computing system through heat transfer associated with the flowing liquid coolant, and installing a heat exchanger at the rear of the chassis to reduce the risk of liquid coolant leakage, Reduce the need for environmental cooling systems.

Benefits of technology

Reduces the possibility of liquid coolant leakage in the computing system, improves cooling efficiency, reduces the demand for the surrounding environment cooling system, and reduces system cost and complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119002641A_ABST
    Figure CN119002641A_ABST
Patent Text Reader

Abstract

Compact after-loading modular heat exchangers and related methods are disclosed herein. An example apparatus disclosed herein includes a heat exchanger having an air flow inlet and an air flow outlet, a cradle frame to receive components of a server chassis, and a bracket coupled to the cradle frame to hold the heat exchanger proximate a rear portion of the server chassis.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates generally to computing components and, more particularly, to compact aftermarket modular heat exchangers and related methods. Background Art

[0002] Due to the increasing need to address thermal management risks caused by increased thermal design power in high-performance systems (such as CPU and / or GPU servers in data centers, cloud computing, edge computing, etc.), the use of liquid cooling of electronic components is being explored to gain advantages over traditional air cooling systems. More specifically, liquids have the inherent advantages of higher specific heat (when boiling is not involved) and higher latent heat of vaporization (when boiling is involved) compared to air. Summary of the invention

[0003] An apparatus is proposed, comprising: a heat exchanger having an air flow inlet and an air flow outlet; a bracket frame for receiving components of a server chassis; and a bracket coupled to the bracket frame for holding the heat exchanger near a rear of the server chassis.

[0004] A corresponding server chassis, method, machine-readable storage medium and device are also proposed. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 One or more example environments are shown in which the teachings of the present disclosure may be implemented.

[0006] Figure 2 At least one example of a data center for executing workloads utilizing disaggregated resources is shown.

[0007] Figure 3 shows that it can be included in Figure 2 At least one example of a pod (ie, distribution point, delivery point, or cluster) in a data center.

[0008] Figure 4 Can be included in Figure 3 A perspective view of at least one example of a rack in a pod.

[0009] Figure 5 yes Figure 4 Side view of the rack.

[0010] Figure 6 A sled is installed in it. Figure 4 A perspective view of the rack.

[0011] Figure 7 yes Figure 6 A block diagram of at least one example of a top side of a skateboard.

[0012] Figure 8 yes Figure 7 A block diagram of at least one example of an underside of a skateboard.

[0013] Fig. 9 is available in Figure 2 A block diagram of at least one example of a compute sled for use in a data center.

[0014] Fig.10 yes Fig. 9 A top perspective view of at least one example of a computing skateboard.

[0015] Fig.11 is available in Figure 2 A block diagram of at least one example of an accelerator sled for use in a data center.

[0016] Fig.12 yes Fig.11 A top perspective view of at least one example of an accelerator slide.

[0017] Fig.13 is available in Figure 2 A block diagram of at least one example of a storage sled for use in a data center.

[0018] Fig.14 yes Fig.13 A top perspective view of at least one example of a storage slide.

[0019] Fig.15 is available in Figure 2 A block diagram of at least one example of a memory sled for use in a data center.

[0020] Fig.16 Yes, you can Figure 2 Block diagram of a system established within a data center to execute workloads using managed nodes with disaggregated resources.

[0021] Fig.17 It is a stereogram of a chassis of the prior art.

[0022] Fig.18 is a perspective view of a chassis including a rear-mounted compact heat exchanger implemented in accordance with the teachings of the present disclosure.

[0023] Fig.19 Yes Fig.18 Exploded view of the heat exchanger and piping assembly.

[0024] Fig. 20 is through Fig.18 The chassis power supply and Fig.17 Schematic diagram of the fluid flow paths of a heat exchanger.

[0025] Figure 21-23 It is a three-dimensional diagram of the chassis in various assembly states.

[0026] Fig.24 is connected to Fig. 22 Power supply Fig.19 A perspective view of a heat exchanger.

[0027] Fig.25 yes Fig.18 A perspective view of a chassis including a replaceable after-installed compact heat exchanger implemented according to the teachings of the present disclosure.

[0028] Fig.26 Yes Fig.18 A perspective view of a liquid coolant distribution system for a chassis and heat exchanger including example cooling system controller circuitry.

[0029] Fig. 27 yes Fig.26 A block diagram of an example implementation of a cooling system controller circuit.

[0030] Fig.28A and Fig.28B is a representation that can be performed, instantiated and / or executed by an example programmable circuit to implement Fig. 27 Example machine readable instructions and / or a flow chart of example operations for a cooling system controller circuit.

[0031] Fig.29 is a block diagram of an example processing platform including programmable circuitry configured to execute, instantiate and / or perform example machine-readable instructions and / or perform Fig.28A and 28B Example operation to achieve Fig. 27 Cooling system controller circuit.

[0032] Fig.30 yes Fig.29 A block diagram of an example implementation of a programmable circuit.

[0033] Fig.31 yes Fig.29 A block diagram of another example implementation of a programmable circuit.

[0034] Generally, the same reference numbers will be used throughout the drawings and accompanying written description to refer to the same or like parts.The drawings are not necessarily drawn to scale.

[0035] As used herein, unless otherwise specified, the term "above" describes the relationship of two components relative to the earth. A first component is above a second component if the second component has at least one component between the earth and the first component. Likewise, as used herein, a first component is "below a second component" when the first component is closer to the earth / ground than the second component. As described above, a first component can be above or below a second component if one or more of the following: other components are between them, no other components are between them, the first and second components are in contact, or the first and third components are not in direct contact with each other.

[0036] As used in this patent, stating that any component (e.g., a layer, film, area, region, or plate) is positioned on another component in any manner (e.g., positioned on, located on, disposed on, or formed on, etc.) means that the referenced component is in contact with the other component, or that the referenced component is on the other component with one or more intermediate components located therebetween.

[0037] As used herein, unless otherwise specified, connection references (e.g., attached, coupled, connected, and joined) may include intermediate members between elements referenced by the connection reference and / or relative movement between those elements. Thus, connection references do not necessarily infer that two elements are directly connected and / or connected in a fixed relationship to each other. As used herein, stating that any component is "in contact" with another component is defined to mean that there are no intermediate components between the two components.

[0038] Unless otherwise specifically stated, descriptors such as "first," "second," "third," etc., are used herein without in any way imposing or otherwise indicating any meaning of priority, physical order, arrangement in a list, and / or ordering, but are merely used as labels and / or arbitrary names to distinguish elements to facilitate understanding of the disclosed examples. In some examples, the descriptor "first" may be used to refer to an element in the detailed description, while a different descriptor such as "second" or "third" may be used in the claims to refer to the same element. In such cases, it should be understood that such descriptors are only used to clearly identify, for example, those elements that might otherwise share the same name in the context of the discussion (e.g., in the claims).

[0039] As used herein, "approximately" and "about" modify the subject matter / values ​​thereof to recognize that there are potential variations that occur in real-world applications. For example, "approximately" and "approximately" may modify dimensions that may not be exact due to manufacturing tolerances and / or other real-world imperfections that would be understood by one of ordinary skill in the art. For example, "approximately" and "approximately" may indicate that such dimensions may be within a tolerance range of + / - 10%, unless otherwise specified in the following description.

[0040] As used herein, "substantially real time" means occurring in a near-instantaneous manner, recognizing that there may be delays in computing time, transmission, etc. in the real world. Therefore, unless otherwise specified, "substantially real time" means real time + 1 second. In some examples used herein, the term "substantially" is used to describe a geometric relationship between two components that is within three degrees of the relationship (e.g., a substantially collinear relationship is within three degrees of linearity, a substantially perpendicular relationship is within three degrees of perpendicularity, a substantially parallel relationship is within three degrees of parallelism, etc.). Similarly, as used herein, a first quantity is "substantially equal" to a second quantity when the first quantity is within 5% of the second quantity.

[0041] As used herein, the phrase "in communication / in a communication manner" includes variations thereof, covers direct communication and / or indirect communication through one or more intermediate components, and does not require direct physical (e.g., wired) communication and / or constant communication, but also includes selective communication at periodic intervals, scheduled intervals, non-periodic intervals and / or one-time events.

[0042] As used herein, "programmable circuit" is defined to include (i) one or more special-purpose circuits (e.g., an application-specific circuit (ASIC)) that are constructed to perform one or more specific operations and include one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors), and / or (ii) one or more general-purpose semiconductor-based circuits that can be programmed with instructions to perform one or more specific functions / one or more specific operations and include one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of programmable circuits include programmable microprocessors, such as a central processing unit (CPU), which can execute a first instruction to perform one or more operations and / or functions; a field programmable gate array (FPGA), which can be programmed with a second instruction to cause the configuration and / or construction of the FPGA to instantiate one or more operations and / or functions corresponding to the first instruction; a graphics processor unit (GPU), which can execute a first instruction to perform one or more operations and / or functions; a digital signal processor (DSP), which can execute a first instruction to perform one or more operations and / or functions; an XPU; a network processing unit (NPU); one or more microcontrollers that can execute a first instruction to perform one or more operations and / or functions; and / or an integrated circuit such as an application specific integrated circuit (ASIC). For example, the XPU can be implemented by a heterogeneous computing system that includes multiple types of programmable circuits (e.g., one or more FPGAs, one or more CPUs, one or more GPUs, one or more NPUs, one or more DSPs, etc., and / or any one or more combinations thereof), and an orchestration technology (e.g., one or more application programming interfaces (APIs)) that can assign one or more computing tasks to multiple types of programmable circuits that are suitable and can be used to perform one or more computing tasks.

[0043] As used herein, an integrated circuit / circuit is defined as one or more semiconductor packages containing one or more circuit elements, such as transistors, capacitors, inductors, resistors, current paths, diodes, etc. For example, an integrated circuit may be implemented as one or more of an ASIC, an FPGA, a chip, a microchip, a programmable circuit, a semiconductor substrate coupling multiple circuit elements, a system on a chip (SoC), etc. DETAILED DESCRIPTION

[0044] As described above, the use of liquid to cool electronic components is explored for its advantages over more traditional air cooling systems because in high-performance systems (e.g., CPU and / or GPU servers in data centers, accelerators, artificial intelligence computing, machine learning computing, cloud computing, edge computing, etc.), there is an increasing need to address thermal management risks caused by increased thermal design power. More specifically, liquids have inherent advantages of higher specific heat (when boiling is not involved) and higher latent heat of vaporization (when boiling is involved) relative to air. In some cases, liquids can be used to indirectly cool electronic components by cooling a cold plate that is thermally coupled to one or more electronic components. Another method is to immerse the electronic components directly in the cooling liquid. In direct immersion cooling, the liquid can be in direct contact with the electronic components to directly take away heat from the electronic components. In order to enable the cooling liquid to be in direct contact with the electronic components, the cooling liquid is electrically insulating (e.g., a dielectric liquid).

[0045] The liquid cooling system may include at least one of single-phase cooling or two-phase cooling. As used herein, single-phase cooling (e.g., single-phase immersion cooling) refers to a cooling fluid (sometimes also referred to herein as a cooling liquid or coolant) used to cool electronic components taking away heat from a heat source (e.g., electronic components) without changing phase (e.g., without boiling and becoming vapor). Such cooling fluids are referred to herein as single-phase cooling fluids, liquids, or coolants. In contrast, as used herein, two-phase cooling (e.g., two-phase immersion cooling) refers to a cooling fluid (in this case, a cooling liquid) evaporating or boiling from the heat generated by the electronic components to be cooled, thereby changing from a liquid phase to a gas phase. The gaseous vapor can then be condensed back into a liquid (e.g., via a condenser) to be used again in the cooling process. Such cooling fluids are referred to herein as two-phase cooling fluids, liquids, or coolants. It is noteworthy that gases (e.g., air) can also be used to cool components and, therefore, can also be referred to as cooling fluids and / or coolants. However, indirect cooling and immersion cooling typically involve at least one cooling liquid (which may or may not become a gas phase when used). Disclosed herein are example systems, apparatus, and related methods for improving cooling systems and / or related cooling processes.

[0046] In recent years, cold plate-based liquid cooling systems have become increasingly common in computing systems. Cold plate systems cool computing systems via conduction between heat generating components (such as a processor) and a cold plate, wherein the cold plate is cooled via the flow of a liquid coolant through it. The coolant circulates through the cold plate to ensure that heat is continuously transferred from the heat generating components. The power supply of the computing system converts alternating current (e.g., from a municipal power grid, etc.) into direct current for use by the system components. Some current power supplies include cold plate cooling systems. While these systems are effective in cooling power supplies, the manufacturing cost of such cooling systems can be high. In addition, the connections in such cooling systems that facilitate the flow of liquid coolant require relatively large forces to dock the power supply to the computing system, which also increases the cost and complexity of such cooling systems.

[0047] The disclosed examples enable cooling of a computing system's power supply via a rear-mounted compact heat exchanger. The examples disclosed herein are air-to-liquid heat exchangers that cool the hot exhaust of a computing system via heat transfer associated with a liquid coolant flowing therein (e.g., convection, etc.). The examples disclosed herein enable existing air-cooled power supplies to benefit from the efficiency of liquid cooling (e.g., when compared to air cooling, etc.) via an externally mounted heat exchanger. The externally mounted heat exchangers disclosed herein reduce the likelihood of liquid coolant leaks within the computing system. In some examples disclosed herein, the temperature of the air exhausted by the heat exchanger is substantially equal to the temperature of the ambient air of the computing system, which reduces the need for cooling systems associated with the ambient environment (such as an air conditioning system of a data center).

[0048] Figure 1 One or more example environments are shown in which the teachings of the present disclosure may be implemented. Figure 1 One or more example environments may include one or more central data centers 102. The central data center 102 may store a large number of servers, for example, for one or more organizations to use for data processing, storage, etc. Figure 1 As shown, one or more central data centers 102 include a plurality of immersion tanks 104 to facilitate cooling servers and / or other electronic components stored at the one or more central data centers 102. The one or more immersion tanks 104 may provide single-phase cooling or two-phase cooling.

[0049] Figure 1 An example environment may be part of an edge computing system. For example, Figure 1An example environment may include an edge data center or micro data center 106. One or more edge data centers 106 may include, for example, a data center located at the base of a cellular tower. In some examples, one or more edge data centers 106 are located at or near the top of a cellular tower and / or other utility pole. One or more edge data centers 106 include corresponding housing for storing one or more servers, wherein the one or more servers may communicate with, for example, one or more servers stored at one or more central data centers 102, client devices, and / or other computing devices in the edge network. Example housing for one or more edge data centers 106 may include materials forming one or more exterior surfaces that partially or completely protect the contents therein, wherein the protection may include weather protection, hazardous environment protection (e.g., EMI, vibration, extreme temperatures), and / or enable immersion. Example housing may include power circuits to provide power for fixed and / or portable implementations, such as an AC power input, a DC power input, one or more AC / DC or DC / AC converters, power conditioners, transformers, charging circuits, batteries, wired inputs, and / or wireless power inputs. As shown in FIG. Figure 1 As shown, the one or more edge data centers 106 may include one or more immersion tanks 108 to store one or more servers and / or one or more other electronic components located at the one or more edge data centers 106 .

[0050] Figure 1 One or more example environments of may include buildings 110 used for commercial and / or industrial purposes, which store information technology (IT) equipment, for example, in one or more rooms of one or more buildings 110. For example, Figure 1 As shown, the one or more servers 112 can be stored with one or more server racks 114 that support the one or more servers 112 (e.g., in an opening of a slot of the rack 114). In some examples, the one or more servers 112 located at the building 110 include one or more on-premises servers of an edge computing network, where the one or more on-premises servers communicate with one or more remote servers (e.g., one or more servers at one or more edge data centers 106) and / or one or more other computing devices within the edge network.

[0051] Figure 1One or more example environments of the embodiment include one or more content delivery network (CDN) data centers 116. The one or more CDN data centers 116 of the example include one or more servers 118 that cache content such as images, web pages, videos, etc. accessed via user devices. The one or more servers 118 of the CDN data center 116 can be disposed in one or more immersion cooling tanks, such as the immersion tanks 104, 108 shown in conjunction with the data centers 102, 106.

[0052] In some cases, Figure 1 In some examples, data centers 102, 106, 116 and / or one or more buildings 110 include servers and / or other electronic components that are cooled independently of an immersion tank (such as immersion tanks 104, 108) and / or an associated immersion cooling system. That is, in some examples, some or all of the servers and / or other electronic components in data centers 102, 106, 116 and / or one or more buildings 110 may be cooled by air and / or liquid coolant without immersing the servers and / or other electronic components therein. Thus, in some examples, the cooling system may be omitted. Figure 1 The immersion tanks 104 and 108 are provided. Figure 1 The example data centers 102, 106, 116 and / or one or more buildings 110 may correspond to the following in combination Figure 2-16 The example data center 200 described in further detail, implemented thereby and / or adapted thereby.

[0053] Although a certain number of cooling slots and one or more other components are shown in the figures, any number of such components may be present. In addition, the example cooling data centers and / or other structures or environments disclosed herein are not limited to Figure 1 For example, a structure containing the example cooling systems and / or components thereof disclosed herein may have an opening sized to accommodate service personnel, such as Figure 1 The example data center 106 of one or more examples disclosed herein may be smaller (e.g., a "doghouse" enclosure). For example, the size of a structure containing the example cooling systems disclosed herein and / or components thereof may be set so that the access (e.g., the only access) to the interior of the structure is a port for service personnel to enter the structure. In some examples, the size of a structure containing the example cooling systems disclosed herein and / or components thereof is set so that only tools can enter the enclosure because the structure may be supported by a utility pole or radio tower or a larger structure.

[0054] Figure 2An example data center 200 is shown in which decomposed resources can collaboratively execute one or more workloads (e.g., applications on behalf of customers). The data center 200 shown includes multiple platforms 210, 220, 230, 240 (referred to herein as pods), each of which includes one or more rows of racks. Although the data center 200 is shown as having multiple pods, in some examples, the data center 200 can be implemented as a single pod. As described in more detail herein, a rack can accommodate multiple sleds. The sled can be primarily equipped with specific types of resources (e.g., memory devices, data storage devices, accelerator devices, general-purpose programmable circuits), that is, resources that can be logically coupled to form a combined node. Some such nodes can act as servers, for example. In an illustrative example, the sleds in pods 210, 220, 230, 240 are connected to multiple pod switches (e.g., switches that route data communications to and from sleds within a pod). The pod switches are in turn connected to the spine switches 250, which exchange communications between the pods (e.g., pods 210, 220, 230, 240) in the data center 200. In some examples, the sleds can use Intel Omni-Path TM Technology and structure connection. In other examples, the skateboard can be connected to other structures, such as InfiniBand or Ethernet. As described in more detail herein, resources within skateboards in the data center 200 can be assigned to groups (referred to herein as "managed nodes") containing resources from one or more skateboards to be used together in the execution of workloads. Workloads can be executed as if the resources belonging to the managed nodes are located on the same skateboard. Resources in a managed node can belong to skateboards belonging to different racks, or even to different pods 210, 220, 230, 240. In this way, some resources of a single skateboard can be assigned to one managed node, while other resources of the same skateboard are assigned to different managed nodes (for example, a first programmable circuit assigned to one managed node and a second programmable circuit of the same skateboard assigned to a different managed node).

[0055] Data centers including disaggregated resources, such as data center 200, can be used in a variety of environments, such as enterprises, governments, cloud service providers, and communications service providers (e.g., Telco's), and can also be used in a variety of sizes, from large data centers for cloud service providers consuming over 200,000 square feet to single or multi-rack installations for base stations.

[0056] In some examples, the decomposition of resources is achieved by using separate sleds that primarily include a single type of resource (e.g., a compute sled that primarily includes compute resources, a memory sled that primarily includes memory resources). The decomposition of resources in this manner, and the selective allocation and deallocation of the decomposed resources to form managed nodes that are allocated to perform workloads, improves the operation and resource utilization of the data center 200 relative to a typical data center. Such a typical data center includes a hyper-conversion server that contains computing, memory, storage, and possibly additional resources in a single chassis. For example, since a given sled will primarily contain resources of the same specific type, resources of that type can be upgraded independently of other resources. In addition, since different resource types (programmable circuits, storage, accelerators, etc.) typically have different refresh rates, greater resource utilization and reduced total cost of ownership can be achieved. For example, a data center operator can upgrade the programmable circuits of the entire facility by simply replacing a compute sled. In this case, accelerators and storage resources may not be upgraded at the same time, but may be allowed to continue operating until these resources are scheduled for their own refresh. Resource utilization may also increase. For example, if managed nodes are composed based on the needs of the workloads that will be run on them, the resources within the node are more likely to be fully utilized. Such utilization can allow more managed nodes to run in a data center with a given set of resources, or allow a data center expected to run a given set of workloads to be built with fewer resources.

[0057] Reference now Figure 3In the illustrative example, the pod 210 includes a set of rows 300, 310, 320, 330 of racks 340. A single rack in the racks 340 can house a plurality of sleds (e.g., sixteen sleds) and provide power and data connections to the housed sleds, as described in more detail herein. In the illustrative example, the racks are connected to a plurality of pod switches 350, 360. The pod switch 350 includes a set of ports 352 to which the sleds of the racks of the pod 210 are connected, and another set of ports 354 that connect the pod 210 to the spine switch 250 to provide connections to other pods in the data center 200. Similarly, the pod switch 360 includes a set of ports 362 to which the sleds of the racks of the pod 210 are connected, and a set of ports 364 that connect the pod 210 to the spine switch 250. In this way, the use of a pair of switches 350, 360 provides a certain amount of redundancy for the pod 210. For example, if any one of the switches 350, 360 fails, the sleds in the pod 210 can still maintain data communications with the rest of the data center 200 (e.g., sleds of other pods) through the other switch 350, 360. In addition, in the illustrative example, the switches 250, 350, 360 can be implemented as dual-mode optical switches that are capable of routing Ethernet protocol communications carrying Internet Protocol (IP) packets and communications according to a second high-performance link layer protocol (e.g., PCI Express) via an optical signaling medium of an optical fabric.

[0058] It should be understood that any of the other pods 220, 230, 240 (and any additional pods of the data center 200) may be associated with the Figure 3 The pod 210 shown and disclosed is similarly constructed and has similar components thereto (e.g., a given pod may have rows of racks that house multiple sleds as described above). Furthermore, while two pod switches 350, 360 are shown, it should be understood that in other examples, there may be a different number of pod switches, thereby providing even more failover capabilities. In other examples, the arrangement of the pods may differ from Figure 2 and Figure 3 For example, a pod may include multiple groups of racks arranged radially, i.e., the racks are equidistant from the central switch.

[0059] Figure 4-6An example rack 340 of the data center 200 is shown. As shown in the illustrated example, the rack 340 includes two vertically arranged elongated support columns 402, 404. For example, when deployed, the elongated support columns 402, 404 can extend upward from the floor of the data center 200. The rack 340 also includes one or more horizontal pairs 410 of elongated support arms 412 (in Figure 4 The plurality of horizontal pairs are configured to support a slide plate of the data center 200, as described below. One of a pair of elongated support arms 412 extends outwardly from the elongated support column 402, while the other elongated support arm 412 extends outwardly from the elongated support column 404.

[0060] In the illustrative example, at least some of the sleds of the data center 200 are chassis-less sleds. That is, such sleds have a chassis-less circuit board substrate on which physical resources (e.g., programmable circuits, memory, accelerators, storage, etc.) are mounted, as discussed in more detail below. As such, the rack 340 is configured to receive a chassis-less sled. For example, a given pair 410 of elongated support arms 412 defines a sled slot 420 of the rack 340 that is configured to receive a corresponding chassis-less sled. To this end, the elongated support arms 412 include corresponding circuit board guides 430 that are configured to receive the chassis-less circuit board substrate of the sled. The circuit board guides 430 are fixed to or otherwise mounted to the top side 432 of the corresponding elongated support arm 412. For example, in the illustrative example, the circuit board guides 430 are mounted at the distal end of the corresponding elongated support arm 412 relative to the corresponding elongated support columns 402, 404. In order Figure 4-6 For clarity, not every circuit board guide 430 may be referenced in every figure. In some examples, at least some of the sleds include a chassis, and the frame 340 is adapted to receive the chassis.

[0061] The circuit board guide 430 includes inner walls defining a circuit board slot 480 that is configured to receive the chassis-free circuit board substrate of the carriage 500 when the carriage 500 is received in the corresponding carriage slot 420 of the rack 340. To this end, as shown in FIG. Figure 5 As shown in FIG. 5 , a user (or robot) aligns the chassis-free circuit board substrate of the exemplary chassis-free slide 500 with the slide slot 420. The user or robot can then slide the chassis-free circuit board substrate forward into the slide slot 420 so that each side edge 514 of the chassis-free circuit board substrate is received in a corresponding circuit board slot 480 of the circuit board guide 430 of the pair 410 of elongated support arms 412 defining the corresponding slide slot 420, as shown in FIG. Figure 5As shown. By having automatically accessible and automatically manipulable sleds that include decomposed resources, different types of resources can be upgraded independently of each other and at their own optimized refresh rates. In addition, the sleds are constructed to blindly mate with power and data communication cables in the racks 340, thereby enhancing their ability to be quickly removed, upgraded, reinstalled and / or replaced. Therefore, in some examples, the data center 200 can be operated (e.g., executing workloads, performing maintenance and / or upgrades, etc.) without human involvement on the data center floor. In other examples, humans can facilitate one or more maintenance or upgrade operations in the data center 200.

[0062] It should be appreciated that the circuit board guide 430 is double-sided. That is, the circuit board guide 430 includes inner walls that define a circuit board slot 480 on each side of the circuit board guide 430. In this manner, the circuit board guide 430 can support a chassis-less circuit board substrate on both sides. In this way, a single additional elongated support column can be added to the rack 340 to turn the rack 340 into a two-rack solution that can accommodate twice as many chassis as the chassis. Figure 4The number of slide slots 420 shown. The exemplary rack 340 includes seven pairs 410 of elongated support arms 412, which define seven corresponding slide slots 420. The slide slots 420 are configured to receive and support corresponding slides 500 as described above. In other examples, the rack 340 may include additional or fewer pairs 410 of elongated support arms 412 (i.e., additional or fewer slide slots 420). It should be understood that because the slide 500 is chassis-less, the slide 500 may have an overall height different from that of a typical server. Therefore, in some examples, the height of a given slide slot 420 may be less than the height of a typical server (e.g., less than a single rack unit, referred to as "1U"). That is, the vertical distance between the pairs 410 of the elongated support arms 412 may be less than a standard rack unit "1U". In addition, due to the relative reduction in the height of the slide slots 420, in some examples, the overall height of the rack 340 may be less than the height of a conventional rack housing. For example, in some examples, the elongated support columns 402, 404 can have a length of six feet or less. Likewise, in other examples, the rack 340 can have different sizes. For example, in some examples, the vertical distance between the pairs 410 of the elongated support arms 412 can be greater than the standard rack unit "1U". In such examples, the increased vertical distance between the slides allows a larger heat sink to be attached to the physical resource, and allows the use of larger fans (e.g., in the fan array 470 described below) to cool the slide, which in turn can allow the physical resource to operate at an increased power level. In addition, it should be understood that the rack 340 does not include any walls, housings, etc. Instead, the rack 340 is a housingless rack open to the local environment. In some cases, in those cases where the rack 340 forms a row end rack in the data center 200, the end plate can be attached to one of the elongated support columns 402, 404.

[0063] In some examples, various interconnects may be routed upward or downward through the elongated support columns 402, 404. To facilitate such routing, the elongated support columns 402, 404 include inner walls that define an interior chamber in which the interconnects may be located. The interconnects routed through the elongated support columns 402, 404 may be implemented as any type of interconnects, including but not limited to data or communication interconnects that provide a communication connection to the skateboard slot 420, power interconnects that provide power to the skateboard slot 420, and / or other types of interconnects.

[0064] In the illustrative example, the rack 340 includes a support platform on which a corresponding optical data connector (not shown) is mounted. Such an optical data connector is associated with a corresponding slide slot 420 and is configured to mate with the optical data connector of the corresponding slide 500 when the slide 500 is received in the corresponding slide slot 420. In some examples, the optical connection between the components (e.g., slides, racks, and switches) in the data center 200 is achieved by a blind-mate optical connection. For example, a door on a given cable can prevent dust from contaminating the optical fibers within the cable. During the connection to the blind-mate optical connector mechanism, the door is pushed open when the cable end approaches or enters the connector mechanism. Subsequently, the optical fibers within the cable can enter the gel within the connector mechanism, and the optical fibers of one cable contact the optical fibers of another cable within the gel within the connector mechanism.

[0065] The illustrative rack 340 also includes a fan array 470 coupled to the lateral support arms of the rack 340. The fan array 470 includes one or more rows of cooling fans 472 aligned in a horizontal line between the elongated support columns 402, 404. In the illustrative example, the fan array 470 includes a row of cooling fans 472 for different slide slots 420 of the rack 340. As described above, in the illustrative example, the slides 500 do not include any onboard cooling system, and thus, the fan array 470 provides cooling for such slides 500 received in the rack 340. In other examples, some or all of the slides 500 may include an onboard cooling system. Additionally, in some examples, the slides 500 and / or the rack 340 may include and / or incorporate a liquid and / or immersion cooling system to facilitate cooling of one or more electronic components on the slide 500. In the illustrative example, the rack 340 also includes different power supplies associated with different ones of the slide slots 420. A given power supply is secured to one of the pair 410 of elongated support arms 412 defining a corresponding skateboard slot 420. For example, the rack 340 may include a power supply coupled or secured to each of the elongated support arms 412 extending from the elongated support column 402. The given power supply includes a power connector configured to mate with a power connector of the skateboard 500 when the skateboard 500 is received in the corresponding skateboard slot 420. In the illustrative example, the skateboard 500 does not include any onboard power supply, and thus, when mounted to the rack 340, a power supply provided in the rack 340 supplies power to the corresponding skateboard 500. A given power supply is configured to meet the power requirements of its associated skateboard, which may be different for different skateboards. In addition, the power supplies provided in the rack 340 may operate independently of each other. That is, within a single rack, a first power supply that provides power to a computing skateboard may provide a power level that is different from a power level supplied by a second power supply that provides power to an accelerator skateboard. The power supply may be controllable at the board level or the rack level and may be controlled locally by components on the associated board or remotely, such as by another board or an orchestrator.

[0066] Reference now Figure 7 In the illustrative example, the sled 500 is configured to be installed in a corresponding rack 340 of the data center 200, as described above. In some examples, a given sled 500 may be optimized or otherwise configured to perform a specific task, such as a computing task, an acceleration task, a data storage task, etc. For example, the sled 500 may be implemented as follows with respect to Fig. 9 and Fig.10 The computing skateboard 900 discussed below is Fig.11 and Fig.12 The accelerator slide 1100 discussed below is Fig.13 and 14The storage sled 1300 discussed above may alternatively be implemented as a sled optimized or otherwise configured to perform other specialized tasks, such as described below with respect to Fig.15 The memory sled 1500 in question.

[0067] As described above, the illustrative slide 500 includes a chassis-less circuit board substrate 702 that supports various physical resources (e.g., electrical components) mounted thereon. It should be understood that the circuit board substrate 702 is "chassis-less" because the slide 500 does not include a housing or shell. Instead, the chassis-less circuit board substrate 702 is open to the local environment. The chassis-less circuit board substrate 702 can be formed of any material capable of supporting the various electrical components mounted thereon. For example, in the illustrative example, the chassis-less circuit board substrate 702 is formed of a FR-4 glass reinforced epoxy laminate. In other examples, other materials may be used to form the chassis-less circuit board substrate 702.

[0068] As discussed in more detail below, the chassis-less circuit board substrate 702 includes a number of features that improve thermal cooling characteristics of various electrical components mounted on the chassis-less circuit board substrate 702. As discussed, the chassis-less circuit board substrate 702 does not include a housing or enclosure, which can improve airflow over the electrical components of the sled 500 by reducing those structures that may inhibit airflow. For example, because the chassis-less circuit board substrate 702 is not positioned in a separate housing or enclosure, there is no vertically arranged backplane (e.g., the backplane of a chassis) attached to the chassis-less circuit board substrate 702, which may inhibit airflow over the electrical components. In addition, the chassis-less circuit board substrate 702 has a geometry that is configured to reduce the length of the airflow path through the electrical components mounted to the chassis-less circuit board substrate 702. For example, the width 704 of the illustrative chassis-less circuit board substrate 702 is greater than the depth 706 of the chassis-less circuit board substrate 702. In one specific example, the chassis-less circuit board substrate 702 has a width of about 21 inches and a depth of about 9 inches, compared to a typical server having a width of about 17 inches and a depth of about 39 inches. Thus, the airflow path 708 extending from the front edge 710 toward the rear edge 712 of the chassis-less circuit board substrate 702 has a shorter distance relative to a typical server, which can improve the thermal cooling characteristics of the sled 500. In addition, although the Figure 7702, but in this example the various physical resources mounted to the chassis-less circuit board substrate 702 are mounted in corresponding locations such that no two substantially heat-generating electrical components shade each other, as discussed in more detail below. That is, two electrical components that do not generate appreciable heat (i.e., greater than a nominal amount of heat sufficient to adversely affect the cooling of another electrical component) during operation are mounted to the chassis-less circuit board substrate 702 linearly in line with each other along the direction of the airflow path 708 (i.e., along a direction extending from the front edge 710 toward the rear edge 712 of the chassis-less circuit board substrate 702). When one or more electrical components are cooled via liquid (e.g., single-phase or two-phase cooling), the placement and / or structure of the features can be appropriately adapted.

[0069] As described above, the illustrative sled 500 includes one or more physical resources 720 mounted to the top side 750 of the chassis-less circuit board substrate 702. Figure 7 500, but it should be understood that in other examples, the sled 500 may include one, two, or more physical resources 720. The physical resources 720 may be implemented as any type of programmable circuit, controller, or other computing circuitry capable of performing various tasks such as computing functions and / or controlling the functionality of the sled 500, depending on, for example, the type or intended functionality of the sled 500. For example, as discussed in more detail below, the physical resources 720 may be implemented as a high-performance processor circuit in examples where the sled 500 is implemented as a computing sled, as an accelerator coprocessor circuit or circuit in examples where the sled 500 is implemented as an accelerator sled, as a storage controller in examples where the sled 500 is implemented as a storage sled, or as a set of memory devices in examples where the sled 500 is implemented as a memory sled.

[0070] The sled 500 also includes one or more additional physical resources 730 mounted to the top side 750 of the chassis-less circuit board substrate 702. In the illustrative example, the additional physical resources include a network interface controller (NIC), as discussed in more detail below. Depending on the type and function of the sled 500, in other examples, the physical resources 730 may include additional or other electrical components, circuits, and / or devices.

[0071] The physical resource 720 is communicatively coupled to the physical resource 730 via an input / output (I / O) subsystem 722. The I / O subsystem 722 may be implemented as circuits and / or components to facilitate input / output operations using the physical resource 720, the physical resource 730, and / or other components of the sled 500. For example, the I / O subsystem 722 may be implemented as or otherwise include a memory controller hub, an input / output control hub, an integrated sensor hub, a firmware device, a communication link (e.g., a point-to-point link, a bus link, a wire, a cable, a waveguide, an optical guide, a printed circuit board trace, etc.), and / or other components and subsystems to facilitate input / output operations. In an illustrative example, the I / O subsystem 722 is implemented as or otherwise includes a double data rate 4 (DDR4) data bus or a DDR5 data bus.

[0072] In some examples, the sled 500 may also include a resource-to-resource interconnect 724. The resource-to-resource interconnect 724 may be implemented as any type of communication interconnect capable of facilitating resource-to-resource communication. In an illustrative example, the resource-to-resource interconnect 724 is implemented as a high-speed point-to-point interconnect (e.g., faster than the I / O subsystem 722). For example, the resource-to-resource interconnect 724 may be implemented as a Quick Path Interconnect (QPI), an Ultra Path Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to resource-to-resource communication.

[0073] The skateboard 500 also includes a power connector 740 that is configured to mate with a corresponding power connector of the rack 340 when the skateboard 500 is installed in the corresponding rack 340. The skateboard 500 receives power from the power supply of the rack 340 via the power connector 740 to power the various electrical components of the skateboard 500. That is, the skateboard 500 does not include any local power source (i.e., onboard power supply) to provide power to the electrical components of the skateboard 500. Excluding a local or onboard power supply helps to reduce the overall footprint of the chassis-less circuit board substrate 702, which can increase the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 702, as described above. In some examples, the voltage regulator is placed in a spaced relationship from the programmable circuit 920 (see Fig. 9 ) directly opposite the bottom side 850 of the chassis-free circuit board substrate 702 (see Figure 8 ) and power is routed from the voltage regulator to the programmable circuit 920 through vias extending through the circuit board substrate 702. This configuration provides an increased thermal budget, additional current and / or voltage, and better voltage control relative to a typical printed circuit board where processor power is delivered in part from the voltage regulator through printed circuit traces.

[0074] In some examples, the skateboard 500 may also include a mounting feature 742 that is configured to mate with a mounting arm or other structure of the robot to facilitate the robot to place the skateboard 500 in the rack 340. The mounting feature 742 may be implemented as any type of physical structure that allows the robot to grasp the skateboard 500 without damaging the chassis-less circuit board substrate 702 or the electrical components mounted thereto. For example, in some examples, the mounting feature 742 may be implemented as a non-conductive pad attached to the chassis-less circuit board substrate 702. In other examples, the mounting feature may be implemented as a bracket, a support, or other similar structure attached to the chassis-less circuit board substrate 702. The specific number, shape, size, and / or composition of the mounting features 742 may depend on the design of the robot configured to manage the skateboard 500.

[0075] Reference now Figure 8 In addition to the physical resources 730 mounted on the top side 750 of the chassis-less circuit board substrate 702, the sled 500 also includes one or more memory devices 820 mounted to the bottom side 850 of the chassis-less circuit board substrate 702. That is, the chassis-less circuit board substrate 702 is implemented as a double-sided circuit board. The physical resources 720 are communicatively coupled to the memory devices 820 via the I / O subsystem 722. For example, the physical resources 720 and the memory devices 820 can be communicatively coupled through one or more vias extending through the chassis-less circuit board substrate 702. In some examples, different ones of the physical resources 720 can be communicatively coupled to different sets of one or more memory devices 820. Alternatively, in other examples, different physical resources 720 can be communicatively coupled to the same memory device in the memory devices 820.

[0076] The memory device 820 may be implemented as any type of memory device capable of storing data of the physical resource 720 during operation of the sled 500, such as any type of volatile memory (e.g., dynamic random access memory (DRAM), etc.) or non-volatile memory. Volatile memory may be a storage medium that requires power to maintain the state of the data stored by the medium. Non-limiting examples of volatile memory may include various types of random access memory (RAM), such as dynamic random access memory (DRAM) or static random access memory (SRAM). One specific type of DRAM that may be used in the memory module is synchronous dynamic random access memory (SDRAM). In a specific example, the DRAM of the memory component may conform to a standard promulgated by JEDEC, such as JESD79F for DDR SDRAM, JESD79-2F for DDR2 SDRAM, JESD79-3F for DDR3 SDRAM, JESD79-4A for DDR4 SDRAM, JESD209 for low power DDR (LPDDR), JESD209-2 for LPDDR2, JESD209-3 for LPDDR3, and JESD209-4 for LPDDR4. Such a standard (and similar standards) may be referred to as a DDR-based standard, and a communication interface of a storage device that implements such a standard may be referred to as a DDR-based interface.

[0077] In one example, the memory device is a block addressable memory device, such as a memory device based on NAND or NOR technology. The memory device may also include a next generation non-volatile device, such as Intel 3D XPoint TMMemory or other byte-addressable write-in-place non-volatile memory device. In one example, the memory device may be or may include a memory device using chalcogenide glass, multi-threshold level NAND flash memory, NOR flash memory, single or multi-level phase change memory (PCM), resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), antiferroelectric memory, magnetoresistive random access memory (MRAM) memory combined with memristor technology, resistive memory including metal oxide substrate, oxygen vacancy substrate and conductive bridge random access memory (CB-RAM) or spin transfer torque (STT)-MRAM, a device based on spintronic magnetic junction memory, a device based on magnetic tunnel junction (MTJ), a device based on DW (domain wall) and SOT (spin-orbit transfer), a thyristor-based memory device, or a combination of any of the above devices, or other memory. The memory device may refer to the die itself and / or the packaged memory product. In some examples, a memory device may include a transistor-stack-free cross-point architecture in which memory cells are located at the intersections of word lines and bit lines and are individually addressable, and in which bit storage is based on changes in bulk resistance.

[0078] Reference now Fig. 9 In some examples, sled 500 may be implemented as a computing sled 900. The computing sled 900 is optimized or otherwise configured to perform computing tasks. As described above, the computing sled 900 may rely on other sleds such as an acceleration sled and / or a storage sled to perform such computing tasks. The computing sled 900 includes various physical resources (e.g., electrical components) similar to the physical resources of the sled 500, which have been described in detail in the prior art. Fig. 9 The same reference numerals are used to identify the above. Figure 7 and Figure 8 The descriptions provided for such components are applicable to corresponding components of computing sled 900 and are not repeated here for clarity of the description of computing sled 900 .

[0079] In the illustrative computing sled 900, the physical resources 720 include programmable circuits 920. Although Fig. 9Only two blocks of programmable circuits 920 are shown in the figure, but it should be understood that in other examples, the computing skateboard 900 may include additional programmable circuits 920. Exemplarily, the programmable circuits 920 correspond to high-performance processor circuits 920 and can be configured to operate at a relatively high power rating. Although the high-performance programmable circuits 920 generate additional heat when operating at a larger power rating than a typical processor circuit (which operates at approximately 155-230W), the enhanced thermal cooling characteristics of the chassis-free circuit board substrate 702 discussed above facilitate higher power operation. For example, in an illustrative example, the programmable circuits 920 are configured to operate at a rated power of at least 250W. In the following example, the programmable circuits 920 can be configured to operate at a rated power of at least 350W.

[0080] In some examples, the computing sled 900 may also include a programmable circuit to programmable circuit interconnect 942. Similar to the resource to resource interconnect 724 of the sled 500 discussed above, the programmable circuit to programmable circuit interconnect 942 may be implemented as any type of communication interconnect capable of facilitating programmable circuit to programmable circuit interconnect 942 communication. In an illustrative example, the programmable circuit to programmable circuit interconnect 942 is implemented as a high-speed point-to-point interconnect (e.g., faster than the I / O subsystem 722). For example, the programmable circuit to programmable circuit interconnect 942 may be implemented as a Quick Path Interconnect (QPI), an Ultra Path Interconnect (UPI), or other high-speed point-to-point interconnect dedicated for programmable circuit to programmable circuit communication.

[0081] The computing sled 900 also includes a communication circuit 930. The illustrative communication circuit 930 includes a network interface controller (NIC) 932, which may also be referred to as a host structure interface (HFI). The NIC 932 may be implemented as or otherwise include any type of integrated circuit, discrete circuit, controller chip, chipset, add-in board, daughter card, network interface card, or other device that can be used by the computing sled 900 to connect to another computing device (e.g., with other sleds 500). In some examples, the NIC 932 may be implemented as part of a system on a chip (SoC) including one or more processor circuits, or included in a multi-chip package that also includes one or more processor circuits. In some examples, the NIC 932 may include a local processor circuit (not shown) and / or a local memory (not shown) that is local to the NIC 932. In such an example, the local processor circuit of the NIC 932 may be capable of performing one or more functions of the programmable circuit 920. Additionally or alternatively, in such an example, the local memory of the NIC 932 may be integrated into one or more components of the computing sled at the board level, the slot level, the chip level, and / or other levels.

[0082] The communication circuit 930 is communicatively coupled to an optical data connector 934. The optical data connector 934 is configured to mate with a corresponding optical data connector of the rack 340 when the compute sled 900 is installed in the rack 340. Illustratively, the optical data connector 934 includes a plurality of optical fibers that are directed from a mating surface of the optical data connector to an optical transceiver 936. The optical transceiver 936 is configured to convert incoming optical signals from the rack-side optical data connector to electrical signals, and convert electrical signals to outgoing optical signals to the rack-side optical data connector. Although shown as forming part of the optical data connector 934 in the illustrative example, in other examples, the optical transceiver 936 may form part of the communication circuit 930.

[0083] In some examples, the computing sled 900 may also include an expansion connector 940. In such an example, the expansion connector 940 is configured to mate with a corresponding connector of an extended chassis-free circuit board substrate to provide additional physical resources to the computing sled 900. For example, during operation of the computing sled 900, the programmable circuit 920 may use the additional physical resources. The extended chassis-free circuit board substrate may be substantially similar to the chassis-free circuit board substrate 702 discussed above, and may include various electrical components mounted thereon. The specific electrical components mounted to the extended chassis-free circuit board substrate may depend on the intended function of the extended chassis-free circuit board substrate. For example, the extended chassis-free circuit board substrate may provide additional computing resources, memory resources, and / or storage resources. Therefore, the additional physical resources of the extended chassis-free circuit board substrate may include, but are not limited to, processor circuits, memory devices, storage devices, and / or accelerator circuits, including, for example, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), security coprocessor circuits, graphics processing units (GPUs), machine learning circuits, or other dedicated processor circuits, controllers, devices, and / or circuits.

[0084] Reference now Fig.10 , which shows an illustrative example of a computing sled 900. As shown, programmable circuits 920, communication circuits 930, and optical data connectors 934 are mounted to the top side 750 of the chassis-less circuit board substrate 702. Any suitable attachment or mounting technique can be used to mount the physical resources of the computing sled 900 to the chassis-less circuit board substrate 702. For example, various physical resources can be mounted in corresponding slots (e.g., processor circuit slots), brackets, or brackets. In some cases, some electrical components can be mounted directly to the chassis-less circuit board substrate 702 via welding or similar techniques.

[0085] As described above, separate programmable circuitry 920 and communication circuitry 930 are mounted to the top side 750 of the chassis-less circuit board substrate 702 such that no two heat-generating electrical components are shielded from each other. In the illustrative example, the programmable circuitry 920 and communication circuitry 930 are mounted to corresponding locations on the top side 750 of the chassis-less circuit board substrate 702 such that no two of these physical resources are linearly aligned with the other physical resources along the direction of the airflow path 708. It should be understood that although the optical data connector 934 is aligned with the communication circuitry 930, the optical data connector 934 does not generate heat or generates nominal heat during operation.

[0086] The memory devices 820 of the computing sled 900 are mounted to the bottom side 850 of the chassis-less circuit board substrate 702, as discussed above with respect to the sled 500. Although mounted to the bottom side 850, the memory devices 820 are communicatively coupled to the programmable circuits 920 located on the top side 750 via the I / O subsystem 722. Because the chassis-less circuit board substrate 702 is implemented as a double-sided circuit board, the memory devices 820 and the programmable circuits 920 can be communicatively coupled through one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 702. In some examples, different programmable circuits 920 (e.g., different processor circuits) can be communicatively coupled to different sets of one or more memory devices 820. Alternatively, in other examples, different programmable circuits 920 (e.g., different processor circuits) can be communicatively coupled to the same memory device in the memory devices 820. In some examples, the memory devices 820 can be mounted to one or more memory mezzanines on the bottom side of the chassis-less circuit board substrate 702 and can be interconnected with corresponding programmable circuits 920 through a ball grid array.

[0087] Different programmable circuits 920 (e.g., different processor circuits) include corresponding heat sinks 950 secured thereto and / or associated therewith. Because the memory devices 820 are mounted to the bottom side 850 of the chassis-less circuit board substrate 702 (and the vertical spacing of the slide 500 in the corresponding rack 340), the top side 750 of the chassis-less circuit board substrate 702 includes additional "free" area or space that facilitates the use of heat sinks 950 having larger sizes relative to conventional heat sinks used in typical servers. In addition, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 702, none of the programmable circuit heat sinks 950 include a cooling fan attached thereto. That is, the heat sink 950 can be a fanless heat sink. In some examples, the heat sink 950 mounted on top of the programmable circuit 920 can overlap with a heat sink attached to the communication circuit 930 in the direction of the airflow path 708 due to its increased size, such as Fig.10 shown.

[0088] Reference now Fig.11 In some examples, sled 500 may be implemented as accelerator sled 1100. Accelerator sled 1100 is configured to perform specialized computational tasks, such as machine learning, encryption, hashing, or other computationally intensive tasks. In some examples, for example, computation sled 900 may offload tasks to accelerator sled 1100 during operation. Accelerator sled 1100 includes various components similar to those of sled 500 and / or computation sled 900, which are described in detail in detail in the accompanying drawings. Fig.11 The same reference numerals are used to identify the above. Figure 7 , Figure 8 and Fig. 9 The descriptions provided for such components are applicable to corresponding components of the accelerator slide 1100 and are not repeated here for clarity of the description of the accelerator slide 1100 .

[0089] In the illustrative accelerator sled 1100, the physical resources 720 include an accelerator circuit 1120. Although Fig.11 Only two accelerator circuits 1120 are shown in FIG. 1 , but it should be understood that in other examples, the accelerator sled 1100 may include additional accelerator circuits 1120. For example, Fig.12 As shown in , the accelerator sled 1100 may include four accelerator circuits 1120. The accelerator circuits 1120 may be implemented as any type of processor circuit, coprocessor circuit, computational circuit, or other device capable of performing computational or processing operations. For example, the accelerator circuits 1120 may be implemented as, for example, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a security coprocessor circuit, a graphics processing unit (GPU), a neuromorphic processor unit, a quantum computer, a machine learning circuit, or other application specific processor circuit, controller, device, and / or circuit.

[0090] In some examples, the accelerator sled 1100 may also include an accelerator-to-accelerator interconnect 1142. Similar to the resource-to-resource interconnect 724 of the sled 500 discussed above, the accelerator-to-accelerator interconnect 1142 may be implemented as any type of communication interconnect that can facilitate accelerator-to-accelerator communication. In an illustrative example, the accelerator-to-accelerator interconnect 1142 is implemented as a high-speed point-to-point interconnect (e.g., faster than the I / O subsystem 722). For example, the accelerator-to-accelerator interconnect 1142 may be implemented as a fast path interconnect (QPI), an ultra-path interconnect (UPI), or other high-speed point-to-point interconnects dedicated to programmable circuit to programmable circuit communication. In some examples, the accelerator circuit 1120 may be daisy-chained with a main accelerator circuit 1120 connected to the NIC 932 and the memory 820 through the I / O subsystem 722 and an auxiliary accelerator circuit 1120 connected to the NIC 932 and the memory 820 through the main accelerator circuit 1120.

[0091] Reference now Fig.12 , which shows an illustrative example of an accelerator sled 1100. As described above, the accelerator circuit 1120, the communication circuit 930, and the optical data connector 934 are mounted to the top side 750 of the chassis-less circuit board substrate 702. Also as described above, the separate accelerator circuit 1120 and the communication circuit 930 are mounted to the top side 750 of the chassis-less circuit board substrate 702 so that no two heat-generating electrical components are shielded from each other. The memory device 820 of the accelerator sled 1100 is mounted to the bottom side 850 of the chassis-less circuit board substrate 702, as discussed above with respect to the sled 500. Although mounted to the bottom side 850, the memory device 820 is communicatively coupled to the accelerator circuit 1120 located on the top side 750 via the I / O subsystem 722 (e.g., through-holes). In addition, the accelerator circuit 1120 may include and / or be associated with a heat sink 1150 that is larger than conventional heat sinks used in servers. As described above with reference to Fig. 9 As discussed above with respect to heat sink 950 , heat sink 1150 may be larger than conventional heat sinks because the “empty” area provided by memory resources 820 is located on the bottom side 850 of chassis-less circuit board substrate 702 rather than on the top side 750 .

[0092] Reference now Fig.13In some examples, the sled 500 may be implemented as a storage sled 1300. The storage sled 1300 is configured to store data in a data store 1350 local to the storage sled 1300. For example, during operation, the computing sled 900 or the accelerator sled 1100 may store data and retrieve data from the data store 1350 of the storage sled 1300. The storage sled 1300 includes various components similar to those of the sled 500 and / or the computing sled 900, which are described in detail in detail in the accompanying drawings. Fig.13 The same reference numerals are used to identify the above. Figure 7 , Figure 8 and Fig. 9 The descriptions provided for such components are applicable to the corresponding components of the storage slide 1300 and are not repeated here for clarity of the description of the storage slide 1300.

[0093] In the illustrative storage sled 1300, the physical resources 720 include a storage controller 1320. Although Fig.13 Only two storage controllers 1320 are shown in the figure, but it should be understood that in other examples, the storage sled 1300 may include additional storage controllers 1320. The storage controllers 1320 may be implemented as any type of programmable circuit, controller, or control circuit that is capable of controlling the storage and retrieval of data into the data storage 1350 based on requests received via the communication circuit 930. In the illustrative example, the storage controllers 1320 are implemented as relatively low-power programmable circuits or controllers. For example, in some examples, the storage controllers 1320 may be configured to operate at a rated power of approximately 75 watts.

[0094] In some examples, the storage sled 1300 may also include a controller-to-controller interconnect 1342. Similar to the resource-to-resource interconnect 724 of the sled 500 discussed above, the controller-to-controller interconnect 1342 may be implemented as any type of communication interconnect capable of facilitating controller-to-controller communication. In an illustrative example, the controller-to-controller interconnect 1342 is implemented as a high-speed point-to-point interconnect (e.g., faster than the I / O subsystem 722). For example, the controller-to-controller interconnect 1342 may be implemented as a Quick Path Interconnect (QPI), an Ultra Path Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to programmable circuit to programmable circuit communication.

[0095] Reference now Fig.14, which shows an illustrative example of a storage sled 1300. In the illustrative example, the data storage 1350 is implemented as or otherwise includes a storage cage 1352 configured to accommodate one or more solid state drives (SSDs) 1354. To this end, the storage cage 1352 includes a plurality of mounting slots 1356 that are configured to receive corresponding solid state drives 1354. The mounting slots 1356 include a plurality of drive guides 1358 that cooperate to define access openings of the corresponding mounting slots 1356. The storage cage 1352 is secured to the chassis-less circuit board substrate 702 such that the access opening faces away from (i.e., toward the front of) the chassis-less circuit board substrate 702. In this way, the solid state drives 1354 are accessible when the storage sled 1300 is installed in the corresponding rack 340. For example, the solid state drives 1354 can be replaced out of the rack 340 (e.g., via a robot) while the storage sled 1300 remains installed in the corresponding rack 340.

[0096] The storage cage 1352 illustratively includes sixteen mounting slots 1356 and is capable of mounting and storing sixteen solid-state drives 1354. In other examples, the storage cage 1352 can be configured to store additional or fewer solid-state drives 1354. Furthermore, in the illustrative example, the solid-state drives are mounted vertically in the storage cage 1352, but may be mounted in a different orientation in the storage cage 1352 in other examples. A given solid-state drive 1354 may be implemented as any type of data storage device capable of storing long-term data. To this end, the solid-state drive 1354 may include the volatile and non-volatile memory devices discussed above.

[0097] like Fig.14 As shown, the storage controller 1320, communication circuitry 930, and optical data connectors 934 are illustratively mounted to the top side 750 of the chassis-less circuit board substrate 702. Again, as described above, any suitable attachment or mounting technique may be used to mount the electrical components of the storage sled 1300 to the chassis-less circuit board substrate 702, including, for example, sockets (e.g., processor circuit sockets), brackets, brackets, soldered connections, and / or other mounting or securing techniques.

[0098] As described above, the separate storage controller 1320 and communication circuit 930 are mounted to the top side 750 of the chassis-less circuit board substrate 702 such that no two heat-generating electrical components are shielded from each other. For example, the storage controller 1320 and communication circuit 930 are mounted in corresponding locations on the top side 750 of the chassis-less circuit board substrate 702 such that no two of these electrical components are linearly aligned with each other along the direction of the airflow path 708.

[0099] The memory device 820 (not shown) of the storage slide 1300 Fig.14) is mounted to the bottom side 850 (not shown) of the chassis-less circuit board substrate 702 Fig.14 ), as discussed above with respect to the sled 500. Although mounted to the bottom side 850, the memory device 820 is communicatively coupled to the storage controller 1320 located on the top side 750 via the I / O subsystem 722. Furthermore, because the chassis-less circuit board substrate 702 is implemented as a double-sided circuit board, the memory device 820 and the storage controller 1320 can be communicatively coupled through one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 702. The storage controller 1320 includes and / or is associated with a heat sink 1370 secured thereto. As described above, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 702 of the storage sled 1300, none of the heat sinks 1370 include a cooling fan attached thereto. That is, the heat sink 1370 can be a fanless heat sink.

[0100] Reference now Fig.15 In some examples, the sled 500 may be implemented as a memory sled 1500. The memory sled 1500 is optimized or otherwise configured to provide other sleds 500 (e.g., the compute sled 900, the accelerator sled 1100, etc.) with access to a pool of memory local to the storage sled 1300 (e.g., in two or more sets 1530, 1532 of the memory devices 820). For example, during operation, the compute sled 900 or the accelerator sled 1100 may remotely write to and / or read from one or more of the memory banks 1530, 1532 of the storage sled 1300 using a logical address space mapped to physical addresses in the memory banks 1530, 1532. The memory sled 1500 includes various components similar to those of the sled 500 and / or the compute sled 900, which are described in detail in detail in detail. Fig.15 The same reference numerals are used to identify the above. Figure 7 , Figure 8 and Fig. 9 The descriptions provided for such components are applicable to corresponding components of the memory sled 1500 and are not repeated here for clarity of the description of the memory sled 1500 .

[0101] In the illustrative memory sled 1500, the physical resources 720 include a memory controller 1520. Although Fig.15Only two memory controllers 1520 are shown in FIG. 1 , but it should be understood that in other examples, the memory sled 1500 may include additional memory controllers 1520. The memory controllers 1520 may be implemented as any type of programmable circuit, controller, or control circuit capable of controlling the writing and reading of data to the memory banks 1530, 1532 based on requests received via the communication circuit 930. In the illustrative example, the memory controllers 1520 are connected to the corresponding memory banks 1530, 1532 to write to and read from the memory devices 820 (not shown) within the corresponding memory banks 1530, 1532, and implement any permissions (e.g., read, write, etc.) associated with the sled 500 that has sent a request to perform a memory access operation (e.g., read or write) to the memory sled 1500.

[0102] In some examples, the memory sled 1500 may also include a controller-to-controller interconnect 1542. Similar to the resource-to-resource interconnect 724 of the sled 500 discussed above, the controller-to-controller interconnect 1542 may be implemented as any type of communication interconnect capable of facilitating controller-to-controller communication. In an illustrative example, the controller-to-controller interconnect 1542 is implemented as a high-speed point-to-point interconnect (e.g., faster than the I / O subsystem 722). For example, the controller-to-controller interconnect 1542 may be implemented as a quick path interconnect (QPI), an ultrapath interconnect (UPI), or other high-speed point-to-point interconnect dedicated to programmable circuit-to-programmable circuit communication. Thus, in some examples, the memory controller 1520 may access memory within a memory group 1532 associated with another memory controller 1520 through the controller-to-controller interconnect 1542. In some examples, the scalable memory controller is made of multiple smaller memory controllers (referred to herein as "chiplets") on a memory sled (e.g., memory sled 1500). The chiplets may be interconnected (e.g., using EMIB (Embedded Multi-Die Interconnect Bridge) technology). The combined chiplet memory controller can scale to a relatively large number of memory controllers and I / O ports (e.g., up to 16 memory channels). In some examples, the memory controller 1520 can implement memory interleaving (e.g., one memory address is mapped to memory bank 1530, the next memory address is mapped to memory bank 1532, and a third address is mapped to memory bank 1530, etc.). The interleaving can be managed within the memory controller 1520, or from a CPU socket (e.g., of the compute sled 900) over a network link to the memory banks 1530, 1532, and can improve latency associated with performing memory access operations compared to accessing consecutive memory addresses from the same memory device.

[0103] In addition, in some examples, the memory sled 1500 can be connected to one or more other sleds 500 (e.g., in the same rack 340 or an adjacent rack 340) via waveguides using waveguide connectors 1580. In the illustrative example, the waveguide is a 74 mm waveguide that provides 16 Rx (i.e., receive) channels and 16 Tx (i.e., transmit) channels. In the illustrative example, the different channels are 16 GHz or 32 GHz. In other examples, the frequency can be different. Using waveguides can provide high-throughput access to the memory pool (e.g., memory banks 1530, 1532) to another sled (e.g., a sled 500 in the same rack 340 or an adjacent rack 340 as the memory sled 1500) without increasing the load on the optical data connector 934.

[0104] Reference now Fig.16, a system for executing one or more workloads (e.g., applications) may be implemented according to data center 200. In an illustrative example, system 1610 includes an orchestrator server 1620, which may be implemented as a managed node including a computing device (e.g., programmable circuit 920 on a computing sled 900) executing management software (e.g., a cloud operating environment such as OpenStack), which is communicatively coupled to a plurality of sleds 500 including a number of computing sleds 1630 (e.g., similar to computing sled 900), memory sleds 1640 (e.g., similar to memory sled 1500), accelerator sleds 1650 (e.g., similar to accelerator sled 1100), and storage sleds 1660 (e.g., similar to storage sled 1300). One or more of sleds 1630, 1640, 1650, 1660 may be grouped into a managed node 1670, such as by orchestrator server 1620, to collectively execute a workload (e.g., an application 1632 executed in a virtual machine or container). Managed nodes 1670 may be implemented as components of physical resources 720 (such as programmable circuits 920, memory resources 820, accelerator circuits 1120, or data storage 1350) from the same or different sleds 500. In addition, managed nodes may be established, defined, or "spun up" by orchestrator server 1620 when a workload is to be assigned to a managed node or at any other time, and may exist regardless of whether any workload is currently assigned to the managed node. In an illustrative example, orchestrator server 1620 may selectively allocate and / or de-allocate physical resources 720 from sleds 500 and / or add or remove one or more sleds 500 (e.g., applications 1632) from managed nodes 1670 based on quality of service (QoS) targets (e.g., target throughput, target latency, target number of instructions per second, etc.) associated with a service level agreement for a workload. In doing so, orchestrator server 1620 may receive telemetry data indicating performance conditions (e.g., throughput, latency, instructions per second, etc.) in different sleds in sled 500 of managed node 1670, and compare the telemetry data to the quality of service targets to determine whether the quality of service targets are met. Orchestrator server 1620 may additionally determine whether one or more physical resources may be de-allocated from managed node 1670 while still meeting the QoS targets, thereby freeing those physical resources for use in another managed node (e.g., executing a different workload). Alternatively, if the QoS targets are not currently being met, orchestrator server 1620 may determine to dynamically allocate additional physical resources to assist in the execution of the workload (e.g., application 1632) while the workload is being executed.Similarly, orchestrator server 1620 can determine to dynamically deallocate physical resources from a managed node if orchestrator server 1620 determines that deallocating physical resources will result in QoS objectives still being met.

[0105] Additionally, in some examples, orchestrator server 1620 can identify trends in resource utilization of a workload (e.g., application 1632), such as by identifying an execution phase (e.g., a time period during which different operations with different resource utilization characteristics are executed) of the workload (e.g., application 1632), and preemptively identify available resources in data center 200 and allocate them to managed nodes 1670 (e.g., within a predetermined time period starting from the associated phase). In some examples, orchestrator server 1620 can model performance based on various latency and distribution schemes to place workloads between compute sleds and other resources (e.g., accelerator sleds, memory sleds, storage sleds) in data center 200. For example, orchestrator server 1620 can utilize a model that considers the performance of resources on sled 500 (e.g., FPGA performance, memory access latency, etc.) and the performance of paths through the network to the resources (e.g., FPGA) (e.g., congestion, latency, bandwidth). Thus, orchestrator server 1620 can determine which resources should be used with which workloads based on the total latency associated with one or more different potential resources available in data center 200 (e.g., latency associated with the performance of the resources themselves, in addition to latency associated with the path through the network between the compute sled executing the workload and sled 500 where the resources are located).

[0106] In some examples, orchestrator server 1620 can use telemetry data (e.g., temperature, fan speed, etc.) reported from sled 500 to generate a map of heat generation in data center 200, and allocate resources to managed nodes based on the map of heat generation and predicted heat generation associated with different workloads to maintain target temperature and heat distribution in data center 200. Additionally or alternatively, in some examples, orchestrator server 1620 can organize the received telemetry data into a hierarchical model indicating relationships between managed nodes (e.g., spatial relationships, such as the physical location of resources of managed nodes within data center 200, and / or functional relationships, such as grouping of managed nodes by customers they provide services for, types of functions typically performed by managed nodes, managed nodes that typically share or exchange workloads with each other, etc.). Based on differences in physical location and resources among managed nodes, a given workload can exhibit different resource utilization on resources of different managed nodes (e.g., causing different internal temperatures, using different percentages of programmable circuits or memory capacity). Orchestrator server 1620 can determine the difference based on the telemetry data stored in the hierarchical model and factor the difference into the prediction of future resource utilization of the workload if the workload is reallocated from one managed node to another managed node to accurately balance resource utilization in data center 200. In some examples, orchestrator server 1620 can identify patterns in resource utilization phases of the workload and use these patterns to predict future resource utilization of the workload.

[0107] In order to reduce the computational load on orchestrator server 1620 and the data transmission load on the network, in some examples, orchestrator server 1620 can send self-check information to sleds 500 to enable a given sled 500 to locally (e.g., on sled 500) determine whether the telemetry data generated by sled 500 meets one or more conditions (e.g., available capacity meeting a predefined threshold, temperature meeting a predefined threshold, etc.). The given sled 500 can then report a simplified result (e.g., yes or no) to orchestrator server 1620, which can use the simplified result to determine resource allocation to the managed nodes.

[0108] Fig.17 1 is a perspective view of a chassis 1700 of the prior art. Fig.17 In FIG. 1 , chassis 1700 includes frame 1702, which includes computing components 1704 and cooling system 1706. Fig.17 In FIG. 1 , cooling system 1706 includes a manifold 1708 that receives liquid coolant from an inlet 1710 and discharges liquid coolant from an outlet 1712. Fig.17In the embodiment, the chassis 1700 includes a first slot 1714A, a second slot 1714B, a third slot 1714C, and a fourth slot 1714D. Fig.17 1714C, and the second power supply assembly 1718 is disposed in the fourth slot 1714D. The chassis 1700 may be a server pod (e.g., Figure 2 server pod 210, etc.), server racks (e.g., Figure 3 and / or Figure 4 Components of a server rack 340, etc.) and / or other parts of a server system (e.g., a server farm, etc.).

[0109] Chassis 1700 is a component that houses computing components 1704. Fig.17 , chassis 1700 includes a frame 1702 that structurally / mechanically supports a computing component 1704 and a cooling system 1706. Computing component 1704 may include one or more processor units (e.g., one or more CPUs, one or more GPUs, one or more accelerators, one or more FPGAs, etc.) and / or related computing components (e.g., permanent memory, temporary memory, etc.).

[0110] The operation of computing components 1704 in chassis 1700 generates heat that is dissipated through the operation of cooling system 1706. Cooling system 1706 is a liquid cooling system that includes internal flow paths that distribute liquid coolant to various locations within frame 1702 of chassis 1700. Inlet 1710 and outlet 1712 are ports that allow liquid coolant to enter and exit cooling system 1706. Fig.17 In FIG. 1 , the cooling system 1706 receives relatively cool liquid coolant via an inlet 1710 and discharges (eg, drives off, etc.) relatively hot liquid coolant via an outlet 1712 . Fig.17 17, inlet 1710 and outlet 1712 are disposed on opposite sides of an example rear portion 1720 of chassis 1700. Cooling system 1706 includes one or more cold plates (not shown) that dissipate heat generated by computing components 1704.

[0111] exist Fig.17 In FIG. 1 , power supply components 1716 and 1718 receive AC power from a power supply of chassis 1700, convert the received current into DC power, and provide the DC power to computing component 1704. Fig.17 1714C and the fourth slot 1714D, respectively. The chassis 1700 includes a first slot 1714A and a second slot 1714B, which may allow for additional power supply components for the chassis 1700.

[0112] The operation of the power supply assemblies 1716, 1718 generates heat, which is dissipated by the cooling system 1706. The manifold 1708 provides a liquid coolant that flows through the internal flow paths of the power supply assemblies 1716, 1718 and dissipates heat therefrom via convection. Such internal flow paths of the power supply assemblies 1716, 1718 may be expensive to design and manufacture, which increases the design time and monetary cost of the power supply assemblies 1716, 1718. In addition, the coupling of the internal flow paths of the power supply assemblies 1716, 1718 may increase the docking force required to couple the power supply assemblies 1716, 1718 to the chassis 1700 and / or the cooling system 1706 (e.g., to ensure a sealed connection through which the liquid coolant flows, etc.). Such high docking forces increase the structural requirements for the power supply assemblies 1716, 1718 and the chassis 1700, which may also increase costs. Additionally, the internal flow paths of the power components 1716 , 1718 increase the number of locations where the cooling system 1706 may leak, which may increase service requirements for the chassis 1700 .

[0113] Fig.18 is a perspective view of an example chassis 1800 including a rear-mounted compact heat exchanger implemented according to the teachings of the present disclosure. Fig.18 In the example shown, chassis 1800 includes a frame 1802 including an example computing component 1804 and an example cooling system 1806. Fig.18 In the example shown, the frame 1802 includes an example first slot 1808A, an example second slot 1808B, an example third slot 1808C, and an example fourth slot 1808D disposed on an example rear portion 1809 of the chassis 1800. Fig.18 In the example shown, the cooling system 1806 receives liquid coolant from an example inlet 1810 and discharges the liquid coolant from an example outlet 1812. Fig.18 In the example shown, the example first heat exchanger 1814 is disposed outside and behind the first tank 1808A, and the example second heat exchanger 1816 is disposed outside and behind the second tank 1808B.

[0114] Frame 1802 structurally / mechanically supports computing components 1804 and cooling system 1806. Fig.18 In the example shown, frame 1802 is a rectangular prism. In other examples, frame 1802 can have any other suitable shape (e.g., different prisms, cubes, cylinders, etc.). In some examples, frame 1802 can include one or more features (e.g., fasteners, brackets, holes, slots, etc.) that allow chassis 1800 to be coupled to a rack (e.g., Figure 3 and / or Figure 4In some examples, the frame 1802 may not be present. In some such examples, the teachings of the present disclosure are implemented in conjunction with a chassis-less slide.

[0115] Computing component 1804 may include one or more processor units (e.g., one or more CPUs, one or more GPUs, one or more accelerators, one or more FPGAs, etc.) and / or related computing components (e.g., permanent memory, temporary memory, etc.). Computing component 1804 may be arranged in any suitable configuration, including a shaded configuration (e.g., a shaded form factor, etc.) and / or an extended core configuration (e.g., an extended core form factor, etc.). In some examples, computing component 1804 may have one or more cold plates that allow cooling system 1806 to dissipate heat therefrom via a flow of liquid coolant (e.g., convection, etc.). In other examples, computing component 1804 may be cooled via cooling system 1806 in any other suitable manner.

[0116] exist Fig.18 In the example shown, the example cooling system 1806 is similar to Fig.17 The cooling system 1706 is different in that Fig.18 The example cooling system 1806 does not provide liquid coolant for cooling the power supplies in the chassis 1800. Fig.18 In the example shown, cooling system 1806 is a liquid cooling system (e.g., including an internal flow path for a liquid coolant for cooling computing components 1804 via convection, etc.). In other examples, cooling system 1806 can be implemented via a fully air-cooled system (e.g., cooling system 1806 can include one or more fans, etc.) and / or an immersion cooling system (e.g., a portion of chassis 1800 is immersed in a liquid coolant, etc.). In some examples, the liquid coolant of cooling system 1806 can be any suitable liquid, including water, ethylene glycol solutions, propylene glycol solutions, and dielectric fluids (e.g., mineral oil, castor oil, hexane, silicone oil, distilled water, fluorinated ketones, perfluorinated compounds, etc.).

[0117] exist Fig.18 In the example shown, the frame 1802 and the rear portion 1809 of the chassis 1800 include slots 1808A, 1808B, 1808C, 1808D. The slots 1808A, 1808B, 1808C, 1808D are openings in the frame 1802 that enable one or more power supplies to be connected to the frame 1802. Fig.18In the example shown, the slots 1808A, 1808B, 1808C, 1808D have substantially the same dimensions (e.g., the same length, the same width, the same height, the same depth, etc.). In other examples, some or all of the slots 1808A, 1808B, 1808C, 1808D may have different dimensions (e.g., different lengths, different widths, different heights, different depths, etc.). Fig.18 In the example shown, slots 1808A, 1808B, 1808C, 1808D are aligned along a rear portion 1809 of chassis 1800. In other examples, slots 1808A, 1808B, 1808C, 1808D may have any other suitable configuration. In some examples, slots 1808A, 1808B, 1808C, 1808D are ports of chassis 1800 that may receive one or more corresponding power supplies. Although chassis 1800 is depicted as having four slots (e.g., slots 1806A, 1806B, 1806C, 1806D, etc.), in other examples, chassis 1800 may include any other suitable number of slots (e.g., one slot, two slots, four slots, etc.).

[0118] Inlet 1810 and outlet 1812 are ports that allow liquid coolant to enter and exit cooling system 1806. Fig.18 In the example shown, the inlet 1810 and the outlet 1812 are disposed on opposite sides of the rear 1809 of the chassis 1800. In other examples, the inlet 1810 and the outlet 1812 may be disposed at any other suitable location (e.g., on the rear 1809 of the chassis 1800, on one side of the chassis 1800, at another location on the front of the chassis 1800, etc.). In other examples, the inlet 1810 and / or the outlet 1812 may not be present (e.g., if the cooling system 1806 is an air cooling system, etc.). In some examples, during operation of the chassis 1800 and its associated components, the coolant enters the cooling system 1806 via the inlet and exits the cooling system 1806 via the exhaust port at a constant and equal or substantially constant and equal rate, which enables continuous cooling of the computing components 1804.

[0119] Heat exchangers 1814, 1816 are compact air-to-liquid coolants implemented in accordance with the teachings of the present disclosure. Heat exchangers 1814, 1816 cool the exhaust air of some or all of the power components associated with chassis 1800. Fig.18 In the example shown, the first heat exchanger 1814 is disposed outside and behind the first tank 1808A, and the example second heat exchanger 1816 is disposed outside and behind the second tank 1808B. Fig.18In the example shown, the first heat exchanger 1814 is separate and distinct from the second heat exchanger 1816. In other examples, the first heat exchanger 1814 and the second heat exchanger 1816 can be integrated. Fig.25 An example chassis including an integral heat exchanger implemented according to the teachings of the present disclosure is described.

[0120] Heat exchangers 1814, 1816 receive a flow of hot air (e.g., relatively hot airflow, etc.) moving toward the rear of chassis 1800 (e.g., from a power supply associated with chassis 1800, etc.), cool the airflow, and expel (e.g., exhaust, etc.) the cooled air into the surrounding environment of chassis 1800 (e.g., aisles of a data center, etc.). Fig.18 In the example shown, the first heat exchanger 1814 receives a hot exhaust flow from a power source associated with the first tank 1808A, and the second heat exchanger 1816 receives a hot exhaust flow from a power source associated with the second tank 1808B.

[0121] In some examples, heat exchangers 1814, 1816 include an internal flow path (e.g., a tube, a channel, a conduit, etc.) containing a liquid coolant that absorbs heat from an airflow passing through heat exchangers 1814, 1816. In some such examples, the internal flow path of heat exchangers 1814, 1816 receives liquid coolant from the same source as cooling system 1806 of chassis 1800, and so on. In other examples, the internal flow path of heat exchangers 1814, 1816 can receive liquid coolant from a source different from cooling system 1806. In some examples, the flow of liquid coolant through heat exchangers 1814, 1816 can be controlled via one or more controllable features (e.g., valves, conduits, etc.). In some such examples, first heat exchanger 1814 can have a different liquid coolant flow rate than second heat exchanger 1816. Fig.26 An example control system for regulating the flow of coolant through heat exchangers 1814, 1816 is described below. Fig. 27 Example operations for regulating coolant flow through heat exchangers 1814, 1816 are described.

[0122] In some examples, the interior of the first heat exchanger 1814 and / or the second heat exchanger 1816 includes internal structures (e.g., fins, channels, etc.) that increase the surface area of ​​the heat exchanger to increase the rate of heat transfer between the airflow from the chassis 1800 and the heat exchangers 1814, 1816. In other examples, no such internal structures are present. Fig.19 , Fig. 20 and Fig.24The first heat exchanger 1814 is described in more detail. Figure 21-23 An example process for assembling chassis 1800 and heat exchangers 1814, 1816 is described.

[0123] exist Fig.18 In the example shown, the third slot 1808C and the fourth slot 1808D do not have corresponding heat exchangers coupled thereto. In some such examples, the power supply components associated with the slots 1808C, 1808D can be cooled via circulation of coolant from the cooling system 1806. Additionally or alternatively, the power supply components associated with the slots 1808C, 1808D can be air cooled and hot air can be exhausted at the rear 1809 of the frame 1802 and the chassis 1800. In other examples, the third slot 1808C and the fourth slot 1808D do not include corresponding power supply components. It should be understood that one or both of the third slot 1808C and / or the fourth slot 1808D may include one or more heat exchangers similar to heat exchangers 1814, 1816. Similarly, having Fig.18 Chassis 1800 having different numbers of slots may accommodate a corresponding number of heat exchangers similar to heat exchangers 1814 , 1816 .

[0124] exist Fig.18 In the example shown, chassis 1800 has an example chassis width 1818 and an example chassis height 1820. Fig.18 In the example shown, the first heat exchanger 1814 has an example heat exchanger width 1822 and an example heat exchanger height 1824. Fig.18 In the example shown, the ratio of chassis width 1818 to heat exchanger width 1822 is approximately 6.6. In other examples, the ratio of chassis width 1818 to heat exchanger width 1822 can be any suitable value between 3 and 10. Fig.18 In the example shown, the ratio of chassis height 1820 to heat exchanger height 1824 is approximately 2.25. In other examples, the ratio between chassis height 1820 and heat exchanger height 1824 can be any suitable value between 1 and 4. In some examples, the standardized dimensions of a 1U server chassis have a height of approximately 44.5 mm, a width of approximately 845 mm, and a length of approximately 800 to 950 mm. In some such examples, a heat exchanger implemented in accordance with the teachings of the present disclosure can have a height of approximately 40 mm, a width of approximately 30 mm, and a length of between approximately 75 mm. Fig.18In the example shown, the second heat exchanger 1814 has the same dimensions as the first heat exchanger 1814 (eg, the second heat exchanger 1814 has a heat exchanger width 1822 and a heat exchanger height 1824 , etc.) In other examples, the second heat exchanger 1816 can have different dimensions than the first heat exchanger 1814 .

[0125] Fig.19 Yes Fig.18 An exploded view of an example heat exchanger assembly 1900 of the first heat exchanger 1814 is shown. Fig.19 In the example shown, the first heat exchanger 1814 includes an example body 1902, an example heat exchanger coolant inlet 1904, an example heat exchange coolant outlet 1906, an example first rail 1908A, an example second rail 1908B, an example first side 1910A, and an example second side 1910B. Fig.19 In the example shown, rails 1908A, 1908B define an example first mounting surface 1912A and an example second mounting surface 1912B. Fig.19 In the example shown, the first mounting surface 1912A includes an example first opening 1914A, an example second opening 1914B, and an example third opening 1914C. Fig.19 In the example shown, the first side portion 1910A includes an example first bracket fastener 1916A and an example second bracket fastener 1916B.

[0126] The heat exchanger assembly 1900 also includes an example conduit 1918 having an example flow conduit 1919 defined by an example first sidewall 1920A, an example second sidewall 1920B, an example first plate 1922A, and an example second plate 1922B. Fig.19 In the example shown, side walls 1920A, 1920B define an example conduit inlet 1924 and an example conduit outlet 1926. Fig.19 In the example shown, the first plate 1922A includes an example first hole 1928A, an example second hole 1928B, and an example third hole 1928C, and the second plate 1922B includes an example fourth hole 1928D, an example fifth hole 1928E, and an example sixth hole 1928F. Fig.19 In the example shown, the heat exchanger assembly 1900 includes an example first fastener 1930A, an example second fastener 1930B, an example third fastener 1930C, an example fourth fastener 1930D, an example fifth fastener 1930E, and an example sixth fastener 1930F.

[0127] exist Fig.19 In the example shown, the first heat exchanger 1814 is composed of a body 1902, a side 1910A, and a side 1910B. Fig.19 In the example shown, side portions 1910A, 1910B extend from opposite sides of the body 1902. Fig.19 In the example shown, the body 1902 of the first heat exchanger 1814 is generally shaped as a rectangular prism, and the sides 1910A, 1910B extend from relatively smaller ends of the body 1902. In other examples, the body 1902 can have any other suitable shape (e.g., a different prism, a cylinder, etc.). Fig.19 In the example shown, the body 1902 includes an example air flow inlet 1915A and an example air flow outlet 1915B, which enables exhaust air flow from the duct 1918 to pass through the first heat exchanger 1814. In the example shown, the air flow inlet 1915A is formed on a side of the body 1902 opposite the air flow outlet 1915B.

[0128] The heat exchanger coolant inlet 1904 and the heat exchanger coolant outlet 1906 are ports that allow liquid coolant to enter and exit the internal flow path of the first heat exchanger 1814. Fig.18 In the example shown, the heat exchanger coolant inlet 1904 and the heat exchanger coolant outlet 1906 are on the same side of the body 1902 and are adjacent to the air flow outlet 1915B. Fig.19 In the example shown, the heat exchanger coolant inlet 1904 is disposed on the first side 1910A, and the heat exchanger coolant outlet 1906 is disposed on the second side 1910B. In other examples, the heat exchanger coolant inlet 1904 and the heat exchanger coolant outlet 1906 can be disposed at any other suitable location (e.g., one or more other locations on the sides 1910A, 1910B, on a side of the body 1902, on the rear of the body 1902, etc.). In some examples, during operation of the first heat exchanger 1814, the coolant enters the first heat exchanger 1814 via the heat exchanger coolant inlet 1904 and leaves the first heat exchanger 1814 via the heat exchanger coolant outlet 1906 at a constant and equal or substantially constant and equal rate.

[0129] exist Fig.19In the example shown, the side portions 1910A, 1910B have a radiused interface with the body 1902 and the rails 1908A, 1908B. In other examples, the side portions 1910A, 1910B may have any other suitable interface with the body 1902 and the rails 1908A, 1908B (e.g., chamfers, blanks, etc.). In other examples, the side portions 1910A, 1910B may be flush with the surface of the body 1902. The first rail 1908A separates the body 1902 from the first side portion 1910A, and the second rail 1908B separates the body 1902 from the second side portion 1910B. The rails 1908A, 1908B are mechanical bosses protruding from the body 1902. Fig.19 In the example shown, the rails 1908A, 1908B extend around the entire body 1902. In other examples, the rails 1908A, 1908B may extend only over the top and / or bottom of the body 1902. In some examples, one or both of the rails 1908A, 1908B may not be present.

[0130] In some examples, the body 1902, the side portions 1910A, 1910B, and the rails 1908A, 1908B are integral (e.g., comprised of a single part, etc.). In some such examples, the first heat exchanger 1814 can be manufactured via additive manufacturing, casting, and / or negative machining. In other examples, the body 1902, the side portions 1910A, 1910B, and the rails 1908A, 1908B can be manufactured separately and assembled via one or more fasteners, one or more press fits, one or more welds, one or more chemical adhesives, and / or combinations thereof.

[0131] The first side 1910A includes a first bracket fastener 1916A and an example second bracket fastener 1916B that enables the heat exchanger assembly 1900 to be coupled to a chassis (eg, Fig.18 Chassis 1800, etc.). Fig.19In the example shown, the first bracket fastener 1916A and the example second bracket fastener 1916B can be tightened (e.g., reducing the distance between the head of the corresponding bracket fastener and the side 1910A, etc.) and / or loosened (e.g., increasing the distance between the head of the corresponding bracket fastener and the side 1910A, etc.). In some examples, to couple the heat exchanger assembly 1900 to the chassis, the bracket fasteners 1916A, 1916B can be loosened to facilitate receiving a bracket associated with the chassis between the sides of the heads of the bracket fasteners 1916A, 1916B. In some such examples, after receiving the bracket, the bracket fasteners 1916A, 1916B can be tightened to retain the heat exchanger assembly 1900 on the chassis via a friction interface (e.g., a clamping interface, etc.). Additionally or alternatively, the heat exchanger assembly 1900 can be coupled to the chassis in any other suitable manner. In some examples, second side 1910B can include bracket fasteners similar to bracket fasteners 1916A, 1916B.

[0132] exist Fig.19 In the example shown, the conduit 1918 is a mechanical structure that is configured to be used as a slave chassis (e.g., Fig.18 The flow path of the exhaust gas flow from the chassis 1800, etc.) to the first heat exchanger 1814. Fig.19 In the example shown, the conduit 1918 is a unitary part (e.g., a single piece, etc.). In other examples, the conduit 1918 can be composed of multiple parts (e.g., the plates 1922A, 1922B can be separate components from the flow conduit 1919, etc.). The conduit 1918 can be constructed of metal, heat-resistant plastic, polymer, and / or combinations thereof. In some examples, the conduit 1918 can be insulated to reduce heat transfer from the interior of the flow conduit 1919 to the surrounding environment of the first heat exchanger 1814.

[0133] In the example shown, the duct 1918 defines a flow conduit 1919 that extends between a conduit inlet 1924 (e.g., a duct inlet, etc.) of the duct 1918 and a conduit outlet 1926 (e.g., a duct outlet, etc.) of the duct 1918. In some examples, the exhaust gas flow enters the duct 1918 via the conduit inlet 1924 (e.g., from a chassis, etc.), passes through the flow conduit 1919, and enters the first heat exchanger 1814 via the conduit outlet 1926. Fig.19 In the example shown, due to the geometry of the first side wall 1920A and the second side wall 1920B, the conduit inlet 1924 is larger than the conduit outlet 1926. Fig.19In the example shown, the first sidewall 1920A is generally planar (e.g., flat, straight, etc.) and the second sidewall 1920B includes an example corner 1932, which reduces the flow area of ​​the flow conduit 1919. In the example shown, the relationship of the sidewalls 1920A, 1920B is such that the conduit outlet 1926 has a smaller area than the conduit inlet 1924 (e.g., the conduit outlet 1926 defines a plane having a smaller area than the plane defined by the conduit inlet 1924, etc.). In other examples, the sidewalls 1920A, 1920B can have any other suitable geometry (e.g., the corner 1932 may not be present, the first sidewall 1920A may include a corner similar to the corner 1932, etc.). Fig.19 In the example shown, the sidewalls 1920A, 1920B have the same thickness. In other examples, the first sidewall 1920A and the second sidewall 1920B may have different thicknesses.

[0134] In some examples, the geometry (eg, size, shape, etc.) of the conduit inlet 1924 is based on a chassis (eg, Fig.18 In some examples, the geometry (e.g., size, shape, etc.) of the duct outlet 1926 is based on the geometry of the first heat exchanger 1814. In other examples, the geometry of the duct outlet 1926 and / or the duct inlet 1924 can be based on any other suitable considerations.

[0135] exist Fig.19 In the example shown, the plates 1922A, 1922B extend away from the conduit outlet 1926 and the flow conduit 1919 and enable the conduits to be coupled to the first heat exchanger 1814. Fig.19 In the example shown, the first plate 1922A and the second plate 1922B have the same size, thickness, and shape. In other examples, the first plate 1922A and the second plate 1922B can have different sizes, thicknesses, and / or shapes. Fig.19 In the example shown, plates 1922A, 1922B have the same thickness as sidewalls 1920A, 1920B. In other examples, plates 1922A, 1922B may have a thickness different from first sidewall 1920A and / or second sidewall 1920B. In some examples, one or both of plates 1922A, 1922B may not exist. In some such examples, pipe 1918 and first heat exchanger 1814 may be assembled and / or coupled via any suitable combination of fasteners, press fits, interference fits, chemical adhesives and / or welding. In other examples, pipe 1918 and first heat exchanger 1814 are integral parts.

[0136] exist Fig.19 In the example shown, the plates 1922A, 1922B of the duct 1918 abut the mounting surfaces 1912A, 1912B, respectively, such that the duct outlet 1926 of the flow duct 1919 is adjacent to the air flow inlet 1915A of the first heat exchanger 1814. Fig.19 In the example shown, the rails 1908A, 1908B and the mounting surfaces 1912A, 1912B define corresponding channels for the plates 1922A, 1922B. In some such examples, the rails 1908A, 1908B of the first heat exchanger 1814 facilitate alignment of the conduit 1918 and the first heat exchanger 1814 during assembly of the heat exchanger assembly 1900.

[0137] exist Fig.19 In the example shown, fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F couple the first heat exchanger 1814 to the pipe 1918. For example, the first fastener 1930A is inserted through the first hole 1928A of the pipe 1918 and is received by the first opening 1914A of the first heat exchanger 1814, the second fastener 1930B is inserted through the second hole 1928B of the pipe 1918 and is received by the second opening 1914B of the first heat exchanger 1814, and the third fastener 1930C is inserted through the third hole 1928C of the pipe 1918 and is received by the third opening 1914C of the first heat exchanger 1814. Similarly, the fourth fastener 1930D, the fifth fastener 1930E, and the sixth fastener 1930F can be inserted through the fourth hole 1928D, the fifth hole 1928E, and the sixth hole 1928F, respectively, and can be received by corresponding openings in the bottom of the first heat exchanger 1814 ( Fig.19 not shown).

[0138] exist Fig.19In the example shown, fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F are screws configured to be received via threaded passage of an opening (e.g., openings 1914A, 1914B, 1914C, etc.) of the first heat exchanger 1814. In other examples, fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F can be implemented by any other suitable type of fastener (e.g., bolts, rivets, etc.). In other examples, the opening is not threaded and / or tapped. In some such examples, the interior of the first heat exchanger 1814 may include one or more nuts to receive the corresponding threads of fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F. In the example shown, six fasteners (e.g., fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F, etc.) are used to couple the first heat exchanger 1814 to the pipe 1918. In other examples, a different number of fasteners are used to couple the first heat exchanger 1814 to the pipe 1918 (e.g., one fastener, two fasteners, four fasteners and seven fasteners, ten fasteners, etc.).

[0139] exist Fig.19 In the example shown, a first hole 1928A, a second hole 1928B, and a third hole 1928C are formed in the first plate 1922A, and a fourth hole 1928D, a fifth hole 1928E, and a sixth hole 1928F are formed in the second plate 1922B. Fig.19 In the example shown, holes 1928A, 1928B, 1928C, 1928D, 1928E, 1928F are through holes with countersunk grooves to ensure that the heads of fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F are flush with the top surface of the corresponding plate in plates 1922A, 1922B. In other examples, holes 1928A, 1928B, 1928C, 1928D, 1928E, 1928F can have any other suitable shape (e.g., countersunk holes, chamfers, etc.). Fig.19 In the example shown, the first hole 1928A is axially aligned with the fourth hole 1928D, the second hole 1928B is axially aligned with the fifth hole 1928E, and the third hole 1928C is axially aligned with the sixth hole 1928F. In other examples, the holes 1928A, 1928B, 1928C, 1928D, 1928E, 1928F can have any other suitable configuration and / or spatial relationship.

[0140] Additionally or alternatively, the conduit 1918 and the first heat exchanger 1814 may be coupled together via one or more welds between the plates 1922A, 1922B and the mounting surfaces 1912A, 1912B, one or more chemical adhesives between the plates 1922A, 1922B and the mounting surfaces 1912A, 1912B, one or more additional fasteners, one or more press fits, one or more shrink fits, etc. In some such examples, some or all of the holes 1928A, 1928B, 1928C, 1928D, 1928E, 1928F, fasteners 1930A, 1930B, 1930C, 1930D, 1930E, 1930F, and / or openings 1914A, 1914B, 1914C are absent.

[0141] Fig. 20 is through Fig.18 and 19 Schematic diagram of the fluid flow path of the example power supply assembly 2000 and the example first heat exchanger 1814. Fig. 20 In the example shown, the example power supply assembly 2000 includes an example first internal flow path 2002 through which an example exhaust gas flow 2004 flows during operation of the power supply assembly 2000. Fig. 20 In the example shown, the power supply assembly 2000 includes an example fan 2006. Fig. 20 In the example shown, the first heat exchanger 1814 includes example internal coolant passages 2008 and example fins 2010 .

[0142] Fan 2006 is an electric fan that causes exhaust flow 2004 to flow through internal flow path 2002 of power supply assembly 2000 and into first heat exchanger 1814. As fan 2006 drives the exhaust flow through internal flow path 2002 of power supply assembly 2000, the exhaust flow 2004 increases in temperature as it draws heat from power supply assembly 2000, thereby reducing the temperature of power supply assembly 2000. In some examples, the speed of fan 2006 can be controlled to modify the speed of exhaust flow 2004, which changes (1) the rate of heat transfer between exhaust flow 2004 and power supply assembly 2000 and (2) the rate of heat transfer between exhaust flow 2004 and first heat exchanger 1814. In some such examples, the speed of fan 2006 can be controlled via a controller of first heat exchanger 1814 (e.g., Fig.26 The cooling system controller circuit 2602, etc. is used to control the cooling system. Fig. 20 In the example shown, fan 2006 is an axial flow fan. In other examples, fan 2006 can be a centrifugal fan. Fig. 20In the example shown, the power supply includes a single fan (e.g., fan 2006, etc.). In other examples, the power supply assembly 2000 may include a different number of fans (e.g., two fans, three fans, etc.). In some such examples, the power supply assembly 2000 may not include a fan, and the exhaust flow 2004 may be driven via an externally mounted fan (e.g., a fan associated with the first heat exchanger, etc.), one or more other fans coupled to the chassis housing the power supply assembly 2000, and / or natural convection.

[0143] Internal coolant passage 2008 is in Fig.19 The heat exchanger coolant inlet 1904 and Fig.19 The fluid flow path extends between the heat exchanger coolant outlet 1906. When the liquid coolant flows through the internal coolant passage 2008, the exhaust gas flow 2004 and the body of the first heat exchanger 1814 are cooled via heat transfer (e.g., convection, etc.), thereby reducing the temperature of the first heat exchanger 1814 and the exhaust gas flow 2004 and increasing the temperature of the liquid coolant in the internal coolant passage 2008. Fig. 20 In the example shown, the internal coolant passage 2008 consists of a single continuous tube (e.g., a pipe, etc.) of a constant diameter extending between the heat exchanger coolant inlet 1904 and the heat exchanger coolant outlet 1906. In other examples, the internal coolant passage can have any other suitable size and / or shape (e.g., a prismatic shape, etc.). In other examples, the internal coolant passage 2008 can include one or more branch passages between the heat exchanger coolant inlet 1904 and the heat exchanger coolant outlet 1906. In some examples, the diameter of the internal coolant passage 2008 can vary along the length of the internal coolant passage 2008. In some examples, the internal coolant passage 2008 can be composed of a material having a relatively high thermal conductivity (e.g., copper, aluminum, silver, silicon nitride, etc.). In other examples, the internal coolant passage 2008 can be composed of any other suitable material. In some examples, the body 1902 of the first heat exchanger 1814 may include one or more baffles to support the internal coolant passage 2008.

[0144] In some examples, the flow rate within the internal coolant passage 2008 can be controlled to change the rate of heat transfer (e.g., convection, etc.) between the exhaust gas flow 2004 and the first heat exchanger 1814 and / or the liquid coolant within the internal coolant passage 2008. Fig.26 and 27 An example controller for controlling the speed of coolant through the internal coolant passage 2008 is described below. Fig.28A and Fig.28B Example operations for controlling the velocity of coolant through the internal coolant passage 2008 are described.

[0145] exist Fig. 20 In the example shown, the first heat exchanger 1814 is a shell and tube heat exchanger (e.g., the internal coolant passage 2008 is used as a tube, the flow path of the body 1902 is used as a shell, etc.). In other examples, the first heat exchanger 1814 can be implemented by a double pipe heat exchanger, a plate heat exchanger, a spiral heat exchanger, a coil heat exchanger, etc. Fig. 20 In the example shown, the first heat exchanger 1814 has a cross-flow configuration (e.g., the internal coolant passage 2008 is substantially perpendicular to the flow of the exhaust gas flow 2004, etc.). In other examples, the first heat exchanger 1814 can have a parallel flow configuration and / or a cross-flow configuration. Fig. 20 The example shown is depicted as including three 180 degree curves, but for illustration purposes this geometry is included in Fig. 20 In other examples, the internal coolant passage 2008 can have any other suitable geometric shape for the three-dimensional space of the body 1902 of the first heat exchanger 1814. For example, the internal coolant passage 2008 can include any suitable combination of curves and / or bends in three-dimensional space.

[0146] The fins 2010 are a plurality of internal structures within the body 1902 that increase the surface area of ​​the body 1902 of the first heat exchanger 1814 that is exposed to the exhaust gas flow 2004. The fins 2010 increase the rate of heat transfer (e.g., convection, etc.) between the exhaust gas flow and the body 1902 of the first heat exchanger 1814. In some examples, the fins 2010 and / or other structures within the body 1902 may be configured to increase the turbulence of the exhaust gas flow 2004, which increases the rate of heat transfer (e.g., convection, etc.) between the exhaust gas flow 2004 and the body 1902 of the first heat exchanger 1814. Additionally or alternatively, the body 1902 of the first heat exchanger 1814 may include one or more other suitable structures (e.g., channels, bosses, etc.) having a relatively high surface area to increase the rate of convection between the exhaust gas flow 2004 and the first heat exchanger 1814 and / or the internal coolant passage 2008. The fin 2010 may include one or more openings to enable the internal coolant passage 2008 to extend therethrough.

[0147] In some examples, the fins 2010 may be formed during the formation of other components of the first heat exchanger 1814 (e.g., via additive manufacturing, via casting, via machining, etc.). In some such examples, the fins 2010 may be formed of the same material as the body 1902 of the first heat exchanger 1814. In some examples, the fins 2010 may be manufactured separately and coupled within the body 1902 via one or more fasteners, one or more welds, one or more press fits, one or more shrink fits, one or more chemical adhesives, etc. In some such examples, the fins 2010 may be formed of the same material as the body 1902 and / or a different material with relatively high thermal conductivity (e.g., copper, aluminum, silver, silicon nitride, etc.). In some such examples, the body 1902 may include features (e.g., slots, holes, welding surfaces, etc.) that facilitate coupling of the fins 2010 thereto.

[0148] exist Fig. 20 In the example shown, the fin 2010 extends parallel to the flow of the exhaust stream 2004 between the air flow inlet 1915A and the air flow outlet 1915B. In other examples, the fin 2010 may have a different configuration (e.g., perpendicular to a portion of the flow of the exhaust stream 2004, etc.) and / or may be disposed downstream of the air flow inlet 1915A. In other examples, the fin 2010 is not present. Fig. 20 In the example shown, the fins 2010 are evenly distributed along the height of the first heat exchanger 1814. In other examples, the fins 2010 may have any other suitable distribution. Fig. 20 In the example shown, no conduit is provided between the power supply assembly 2000 and the first heat exchanger 1814. In other examples, a conduit (e.g., Fig.19 A conduit 1918, etc.) may be provided in the flow path of the exhaust flow 2004 to direct the exhaust flow 2004 between the power source and the first heat exchanger 1814.

[0149] Figure 21-23 18 is a perspective view depicting an example process of coupling the first heat exchanger 1814 to the chassis 1800. Figure 21-23 One such process is depicted coupling the first heat exchanger 1814 to the chassis 1800 , but it should be understood that the first heat exchanger 1814 and the chassis 1800 may be assembled in any other suitable manner.

[0150] Fig.21 is in the example first state 2100 Fig.18 A perspective view of the chassis 1800. Fig.21 In the example shown, the example bracket assembly 2102 is disposed at Fig.18 Example of the first groove 1808A. Fig.21In the example shown, the bracket assembly 2102 includes an example bracket 2104, which includes an example first mounting feature 2106A, an example second mounting feature 2106B, an example third mounting feature 2106C, and an example fourth mounting feature 2106D. Fig.21 In the example shown, the bracket assembly 2102 includes an example handle 2108 and an example bracket frame 2110 .

[0151] exist Fig.21 In the example shown, the bracket assembly 2102 is shaped to fit within (e.g., be disposed within, etc.) one or more slots (e.g., first slot 1808A, second slot 1808B, etc.) of the chassis 1800. In other examples, the bracket assembly 2102 can have any other suitable shape. The bracket assembly 2102 is configured to accommodate a power supply (e.g., Fig. 20 Power supply assembly 2000, etc.). Fig.21 In the example shown, the bracket assembly 2102 includes a bracket frame 2110. The bracket frame 2110 is a structure configured to hold a power supply and couple the power supply within a slot (e.g., first slot 1808A, second slot 1808B, etc.) of the chassis 1800. In some examples, the bracket frame 2110 may include one or more features that enable the bracket frame 2110 to be disposed within the chassis 1800. For example, the bracket frame 2110 may include one or more rails, one or more slots, one or more fasteners, one or more holes, etc., which facilitate the coupling of the bracket frame 2110 within the slots 1808A, 1808B, etc. In some examples, the bracket frame 2110 may include one or more features that enable the power supply to be disposed therein. For example, the bracket frame 2110 may include one or more rails, one or more slots, one or more fasteners, one or more holes, etc., which facilitate the coupling of the power supply within the bracket frame 2110. The support frame 2110 can be made of one or more metals (e.g., aluminum, steel, copper, etc.), one or more polymers (e.g., thermosetting plastics, thermoplastics, etc.), one or more ceramics, and / or combinations thereof. In some examples, the support frame 2110 can include one or more electrical interfaces (not shown) that enable power to be transmitted to a power supply disposed therein (e.g., AC input, etc.) and / or from a power supply disposed therein (e.g., DC output, etc.). In some examples, the support frame 2110 can include one or more data interfaces (not shown) that allow communication between the power supply disposed therein and other components of the chassis 1800 and / or an external controller (e.g., Fig.26In some examples, the rack frame 2110 may include one or more sensors that measure the exhaust flow rate through the power supply disposed therein, the temperature of the exhaust flow through the power supply disposed therein, etc.

[0152] exist Fig.21 In the example shown, the bracket assembly 2102 includes a bracket 2104 that is coupled to the bracket frame 2110 at a location remote from the chassis 1800. Fig.21 In the example shown, the bracket 2104 protrudes from the rear 1809 of the chassis 1800, and the bracket frame 2110 is disposed within the chassis 1800. Fig.21 In the example shown, the bracket 2104 is integral with the bottom surface of the support frame 2110. In some examples, the bracket 2104 can be made of the same material as the support frame 2110. In other examples, the bracket 2104 can be a separate component from the support frame 2110. For example, the bracket 2104 can be manufactured separately from the support frame 2110 and coupled to the support frame 2110 via one or more fasteners, one or more chemical adhesives, one or more welds, one or more press fits, etc. The bracket 2104 can be made of any suitable material, including metals, polymers, ceramics, and / or combinations thereof.

[0153] exist Fig.21 In the example shown, the bracket 2104 includes mounting features 2106A, 2106B, 2106C, 2106D, and defines a channel 2107. Fig.21 In the example shown, the channel 2107 has a size and shape (eg, sized to allow the power source to slide through, etc.) that facilitates depositing the power source within the support frame 2110. In other examples, the channel 2107 can have any other suitable shape.

[0154] Mounting features 2106A, 2106B, 2106C, 2106D facilitate coupling the heat exchanger within channel 2107 of bracket 2104 to support assembly 2102 and to chassis 1800. Fig.21 In the example shown, the mounting features 2106A, 2106B, 2106C, 2106D are slots formed in the bracket 2104. Fig.21In the example shown, the mounting features 2106A, 2106B, 2106C, 2106D are straight slots. In other examples, the mounting features 2106A, 2106B, 2106C, 2106D can be implemented by slots including one or more curves (e.g., L-shaped slots, C-shaped slots, etc.). In other examples, the mounting features 2106A, 2106B, 2106C, 2106D can be implemented by any other suitable features (e.g., holes, welding surfaces, spring pins, etc.). Fig.21 In the example shown, bracket 2104 includes four mounting features (e.g., mounting features 2106A, 2106B, 2106C, 2106D, etc.) In other examples, bracket 2104 may include any other suitable number of mounting features (e.g., one mounting feature, two mounting features, five mounting features, etc.).

[0155] The handle 2108 is a mechanical structure that facilitates the operation of the bracket assembly 2102. For example, the handle 2108 facilitates the coupling and / or removal of the bracket assembly 2102 within the first slot 1808A. Fig.21 In the example shown, handle 2108 is a U-shaped member that is coupled to support frame 2110 at two locations. In other examples, handle 2108 can have any other suitable shape and can be coupled to support assembly 2102 at any other suitable location. In some examples, handle 2108 may not be present. In some such examples, support assembly 2102 can include other features that facilitate handling, assembly, and / or removal of support assembly 2102.

[0156] exist Fig.21 In the example shown, bracket assembly 2102 can be coupled within first slot 1808A via example translation 2112. In some such examples, bracket assembly 2102 can be retained within first slot 1808A via one or more fasteners and / or locking mechanisms between bracket assembly 2102 and features of chassis 1800 proximate first slot 1808A. In some examples, a bracket assembly similar to bracket assembly 2102 can be disposed within second slot 1808B to enable a heat exchanger to be coupled thereto.

[0157] Fig. 22 is in Fig.21 Example of a second state 2200 after the first state 2100 Fig.18 A perspective view of the chassis 1800. Fig. 22In the example shown, bracket assembly 2102 is disposed in first slot 1808A, and another example bracket assembly 2202 similar to bracket assembly 2102 has been disposed in second slot 1808B. In some examples, bracket assemblies similar to bracket assemblies 2102, 2202 may be disposed in third slot 1808C and / or fourth slot 1808D. Fig. 22 In the example shown, Fig. 20 The power supply assembly 2000 will be disposed within the bracket assembly 2102. Fig. 22 In the example shown, fan 2006 is aligned with channel 2107 of bracket assembly 2102 such that exhaust air from power supply assembly 2000 flows through channel 2107 .

[0158] exist Fig. 22 In the example shown, power supply assembly 2000 can be coupled within bracket assembly 2102 via example translation 2204. In some such examples, power supply assembly 2000 can be retained within bracket assembly 2102 and / or first slot 1808A via one or more fasteners and / or locking mechanisms between features of bracket assembly 2102 and chassis 1800 proximate first slot 1808A.

[0159] Fig.23 is in Fig. 22 The example third state 2300 after the second state 2200 Fig.18 A perspective view of the chassis 1800. Fig.23 In the example shown, power supply assembly 2000 has been disposed within bracket assembly 2102 in first slot 1808A, and another power supply assembly 2302 similar to power supply assembly 2000 has been disposed within bracket assembly 2202 in second slot 1808B. In some examples, power supplies similar to power supply assemblies 2000, 2302 may be disposed within third slot 1808C and / or fourth slot 1808D. Fig.23 In the example shown, Fig.19 The heat exchanger assembly 1900 will be disposed within the bracket assembly 2102 in the first slot 1808A. Fig.23 In the example shown, the heat exchanger assembly 1900 includes Fig.19 The first bracket fastener 1916A, Fig.19 The example second bracket fastener 1916B, the example third bracket fastener 2304A and the example fourth bracket fastener 2304B.

[0160] To couple the heat exchanger assembly 1900 to the bracket assembly 2102 and the chassis 1800, the heat exchanger assembly 1900 can be moved into the channel 2107 of the bracket 2104 via the example translation 2306. Fig.23In the example shown, translation 2306 aligns bracket fasteners 1916A, 1916B, 2304A, 2304B with mounting features 2106A, 2106B, 2106C, 2106D, respectively. Fig.23 In the example shown, to fasten the heat exchanger assembly 1900 to the bracket assembly 2102, the bracket fasteners 1916A, 1916B, 2304A, 2304B may be fastened, which holds the heat exchanger assembly 1900 in the channel 2107 via a friction interface between the first bracket fastener 1916A and the first mounting feature 2106A, a friction interface between the second bracket fastener 1916B and the second mounting feature 2106B, a friction interface between the third bracket fastener 2304A and the third mounting feature 2106C, and a friction interface between the fourth bracket fastener 2304B and the fourth mounting feature 2106D. The relationship between the heat exchanger assembly 1900 and the power supply assembly 2000 will be described below in conjunction with Fig.24 Described in further detail.

[0161] Fig.24 is in Fig.23 The third state 2300 is connected to Fig. 20 and 22 Power Supply Components 2000 Figure 18-23 A perspective view of the first heat exchanger 1814. Fig.24 In the example shown, the heat exchanger assembly 1900 is Fig.21 The connection between the bracket assembly 2102 (e.g., via bracket fasteners 1916A, 1916B, 2304A, 2304B and mounting features 2106A, 2106B, 2106C, 2106D, etc.) places the duct 1918 adjacent to the power supply assembly 2000 so that the duct inlet 1924 is adjacent to the fan 2006 of the power supply assembly 2000. Fig.24 In the example shown, duct inlet 1924 surrounds fan 2006, and fan 2006 and duct inlet 1924 have substantially the same area. In other examples, duct inlet 1924 and fan 2006 can have other suitable relationships. For example, duct inlet 1924 can be substantially larger than fan 2006 (e.g., extending along the entire length of power supply assembly 2000, etc.).

[0162] exist Fig.24 In the example shown, the first heat exchanger 1814 includes an example first sensor 2402, an example second sensor 2404, an example third sensor 2406, and an example fourth sensor 2408. Fig.24In the example shown, the first sensor 2402 and the second sensor 2404 measure the temperature of the air at the air inlet 1915A and the temperature of the air at the air outlet 1915B, respectively. Fig.24 In the example shown, the third sensor 2406 and the fourth sensor 2408 measure the temperature of the liquid coolant at the coolant inlet 1904 and the temperature of the liquid coolant at the coolant outlet 1906 .

[0163] exist Fig.24 In the example shown, the first sensor 2402 is disposed adjacent to the air flow inlet 1915A. In other examples, the first sensor 2402 may be disposed at another location within the duct 1918 and / or near the fan 2006. Fig.24 In the example shown, the third sensor 2406 and the fourth sensor 2408 are respectively disposed in the coolant inlet 1904 and the coolant outlet 1906. In other examples, the third sensor 2406 may be disposed at another location upstream of the coolant system (e.g., at Fig.26 , in the inlet coolant manifold 2604, in the outlet of the coolant distribution unit, etc.), and / or the fourth sensor 2408 may be disposed at another location downstream in the cooling system (e.g., Fig.26 , in the outlet coolant manifold 2606, in the inlet of the coolant distribution unit, etc.). One or more of the sensors 2402, 2404, 2406, 2408 may be implemented by one or more thermocouples, one or more resistance temperature detectors, one or more thermistors, one or more infrared optical sensors, one or more semiconductor-based sensors, and / or combinations thereof.

[0164] Fig.25 is a perspective view of another example chassis 2500 including another example heat exchanger assembly 2502 implemented in accordance with the teachings of the present disclosure. Fig.25 The chassis 2500 is similar to Fig.18 The chassis 1800 is similar to the chassis 2500, except that the chassis 2500 includes an example heat exchanger assembly 2502. Fig.25 In the example shown, the heat exchanger assembly 2502 includes an example heat exchanger 2504, an example first conduit 2506A, an example second conduit 2506B, an example third conduit 2506C, and an example fourth conduit 2506D. Fig.25 In the example shown, pipes 2506A, 2506B, 2506C, and 2506D are respectively disposed in the example first slot 2508A, the example second slot 2508B, the example third slot 2508C, and the example fourth slot 2508D. Fig.25In the example shown, the heat exchanger 2504 includes an example body 2510, an example air flow outlet 2512, an example first side 2514A, an example second side 2514B, an example liquid coolant inlet 2518, an example liquid coolant outlet 2520, an example first bracket fastener 2522A, and an example second bracket fastener 2522B. Fig.25 In the example shown, the chassis 2500 includes an example chassis coolant inlet 2523 and an example chassis coolant outlet 2524 disposed on an example rear portion 2526 of the chassis 2500 .

[0165] Heat exchanger 2504 is an air-to-liquid heat exchanger that cools the exhaust air flow from the rear of chassis 2500. Fig.25 In the example shown, the heat exchanger 2504 is similar to Figure 18-24 The first heat exchanger 1814 is a shell and tube heat exchanger, except that the heat exchanger 2504 receives exhaust air flow from each of the tanks 2508A, 2508B, 2508C, 2508D. In other examples, the heat exchanger 2504 is implemented by a double pipe heat exchanger, a plate heat exchanger, a spiral heat exchanger, a coil heat exchanger, etc. In some examples, the heat exchanger 2504 can cool the exhaust flow from four power sources (e.g., power supplies, etc.) associated with each of the tanks 2508A, 2508B, 2508C, 2508D.

[0166] In some examples, the interior of the body 2510 of the heat exchanger 2504 may include structures that increase the surface area of ​​the body 2510 of the heat exchanger 2504 (e.g., similar to Fig. 20 In some such examples, the internal structure of the heat exchanger 2504 increases the rate of heat transfer (e.g., convection, etc.) between the heat exchanger 2504 and the exhaust flow. In some examples, the internal structure of the body 2510 can be evenly distributed throughout the body 2510. In other examples, the internal structure can have an appropriate distribution in the body 2510. In some examples, the body 2510 does not include such surface area increasing structure.

[0167] The heat exchanger 2504 includes an internal coolant passage between a liquid coolant inlet 2518 and a liquid coolant outlet 2520 that cools the heat exchanger 2504 and the exhaust flow from the chassis 2500 flowing through the chassis. The internal coolant passage of the heat exchanger 2504 can have any suitable configuration (e.g., a cross-flow configuration, a parallel flow configuration, a counter-flow configuration, etc.). In some examples, the internal coolant passage can be a single continuous conduit (e.g., a tube, a pipe, etc.) between the liquid coolant inlet 2518 and the liquid coolant outlet 2520. In other examples, the internal coolant passage can include one or more branch conduits between the liquid coolant inlet 2518 and the liquid coolant outlet 2520. In some such examples, the branching of the liquid coolant inlet 2518 and the liquid coolant outlet 2520 contributes to a substantially equal outlet flow along the length of the heat exchanger 2504 and a uniform distribution of fresh liquid coolant (e.g., relatively cool liquid coolant, etc.).

[0168] The liquid coolant inlet 2518 and the liquid coolant outlet 2520 are ports that allow liquid coolant to enter and exit the internal flow path of the heat exchanger 2504. Fig.25 In the example shown, the liquid coolant inlet 2518 and the liquid coolant outlet 2520 are disposed on the first side 2514A and the second side 2514B, respectively, and are adjacent to the air outlet 2512. Fig.25 In the example shown, the liquid coolant inlet 2518 is located near the first slot 2508A, while the liquid coolant outlet 2520 is located near the second slot 2508B. Fig.25 In the example shown, the liquid coolant inlet 2518 and the liquid coolant outlet 2520 are disposed on the same side of the body 2510 (e.g., the rear side of the body 2510, etc.) and have substantially the same height relative to the body 2510. In other examples, the liquid coolant inlet 2518 and the liquid coolant outlet 2520 can be disposed at any other suitable location (e.g., one or more other locations on the side portions 2514A, 2514B, on a side of the body 2510, on the rear of the body 2510, etc.). In some examples, during operation of the heat exchanger 2504, the coolant enters the heat exchanger 2504 via the liquid coolant inlet 2518 and leaves the heat exchanger 2504 via the liquid coolant outlet 2520 at a constant and equal or substantially constant and equal rate.

[0169] In some examples, the velocity of the liquid coolant flow within the internal flow path of the heat exchanger 2504 can be controlled to change the rate of heat transfer (e.g., convection, etc.) between the exhaust flow of the chassis 2500 and the heat exchanger 2504 and / or within the liquid coolant within the internal flow path of the heat exchanger 2504. In some examples, the flow rate of the liquid coolant in the heat exchanger 2504 can be controlled so that the flow leaving the heat exchanger 2504 via the air flow outlet 2512 is substantially equal in temperature to the surrounding environment of the heat exchanger 2504 and / or the chassis 2500. Fig.26 and 27 An example controller circuit for controlling the speed of coolant through the heat exchanger 2504 is described below. Fig.28A and Fig.28B Example operations for controlling the rate of coolant passing through heat exchanger 2504 are described.

[0170] exist Fig.25 In the example shown, heat exchanger 2504 receives exhaust air from each of slots 2508A, 2508B, 2508C, 2508D via conduits 2506A, 2506B, 2506C, 2506D. Conduits 2506A, 2506B, 2506C, 2506D direct the exhaust flow from each of slots 2508A, 2508B, 2508C, 2508D to heat exchanger 2504. Each of conduits 2506A, 2506B, 2506C, 2506D is similar to Fig.19 The pipe of 1918, the difference is Fig.25 Ducts 2506A, 2506B, 2506C, 2506D direct exhaust air from chassis 2500 to a single heat exchanger, namely heat exchanger 2504. In some examples, heat exchanger 2504 includes an air flow inlet adjacent to each of ducts 2506A, 2506B, 2506C, 2506D. Ducts 2506A, 2506B, 2506C, 2506D may be arranged in a manner similar to that of Fig.19The pipe 1918 of the first heat exchanger 1814 is coupled to the heat exchanger 2504 in a manner that is fastened to the first heat exchanger 1814 (e.g., via a plurality of screws extending through holes on the top and bottom of the pipes 2506A, 2506B, 2506C, 2506D and extending into the heat exchanger 2504). In other examples, the pipes 2506A, 2506B, 2506C, 2506D can be coupled to the heat exchanger 2504 via one or more other fasteners, one or more welds, one or more chemical adhesives, one or more press fits, one or more shrink fits, and / or combinations thereof. Additionally or alternatively, one or more of the pipes 2506A, 2506B, 2506C, 2506D and the heat exchanger 2504 can be an integral component.

[0171] The bracket fasteners 2522A, 2522B are fasteners provided on the first side portion 2514A. Fig.25 In the example shown, the bracket fasteners 2522A, 2522B are similar to Fig.23 In some examples, the second side portion 2514B may include a bracket fastener 1916A, 1916B, 2304A, 2304B. Fig.25 Bracket fasteners 2522A, 2522B and / or Fig.23 In some examples, to couple the heat exchanger assembly 2502 to the chassis 2500, the bracket fasteners 2522A, 2522B can be coupled to the bracket associated with the fourth slot 2508D (e.g., Fig.24 The corresponding mounting features of the bracket 2104 similar to the bracket, etc.) (e.g., similar to Fig.24 8A). In some such examples, a bracket fastener (not shown) associated with the second side 2514B can be aligned with a corresponding mounting feature associated with the first slot 2508A. In some examples, to fasten the heat exchanger assembly 2502 to the chassis 2500, bracket fasteners 2522A, 2522B and bracket fasteners disposed on the second side 2514B can be fastened, which form a friction interface between such bracket fasteners and corresponding mounting features of the chassis 2500. Additionally or alternatively, the heat exchanger assembly 2502 can be coupled to the chassis 2500 via one or more other fasteners, one or more welds, one or more chemical adhesives, one or more press fits, one or more shrink fits, and / or combinations thereof.

[0172] The chassis coolant inlet 2523 and chassis coolant outlet 2524 are ports that allow liquid coolant to enter and exit the cooling system (not shown) of the chassis 2500. Fig.25 In the example shown, the chassis coolant inlet 2523 and the chassis coolant outlet 2524 are disposed on opposite sides of a rear portion 2526 of the chassis 2500. In other examples, the chassis coolant inlet 2523 and the chassis coolant outlet 2524 can be disposed at any other suitable location (e.g., another location on the rear portion 2526 of the chassis 2500, on a side of the chassis 2500, on the front of the chassis 2500, etc.). In some examples, the chassis coolant inlet 2523 and the liquid coolant inlet 2518 can receive liquid coolant from the same source (e.g., a coolant distribution unit (CDU) associated with the chassis 2500, etc.). In some examples, the liquid coolant outlet 2520 and the chassis coolant outlet 2524 can expel liquid coolant to the same source (e.g., a coolant distribution unit (CDU) associated with the chassis 2500, etc.). In other examples, coolant inlet 2523 and chassis coolant outlet 2524 may not be present (eg, if chassis 2500 has an air cooling system, etc.).

[0173] exist Fig.25 In the example shown, chassis 2500 has an example chassis width 2528 and an example chassis height 2530. Fig.25 In the example shown, the heat exchanger 2504 has an example heat exchanger width 2532 and an example heat exchanger height 2534. Fig.25 In the example shown, the ratio of the chassis width 2528 to the heat exchanger width 2532 is approximately 1.48. In other examples, the ratio of the chassis width 2528 to the heat exchanger width 2532 can be any suitable value between 1 and 3. Additionally or alternatively, the heat exchanger width 2532 is based on the distance between the first slot 2508A and the fourth slot 2508D (e.g., the heat exchanger 2504 is long enough to accommodate the exhaust flow from the first slot 2508A and the fourth slot 2508D, etc.). Fig.25 In the example shown, the ratio of the chassis height 2530 to the heat exchanger height 2534 is approximately 2.25. In other examples, the ratio between the chassis height 2530 and the heat exchanger height 2534 can be any suitable value between 1 and 4.

[0174] Combination Fig.25Heat exchanger assembly 2502 is depicted cooling exhaust streams from four power supplies. It should be understood that the teachings disclosed herein are also applicable to heat exchanger assemblies serving different numbers of slots. For example, a heat exchanger assembly implemented in accordance with the teachings of the present disclosure and similar to heat exchanger assembly 2502 can provide cooling to a chassis having any number of slots (e.g., two slots, three slots, five slots, ten slots, etc.). In addition, a heat exchanger assembly implemented in accordance with the teachings of the present disclosure and similar to Fig.25 The two heat exchanger assemblies of the heat exchanger assembly can be similar to Fig.25 One or more of the chassis 2500 provide exhaust cooling (e.g., (1) a first heat exchanger assembly is capable of cooling the exhaust of the first slot 2508A and the second slot 2508B, while the second heat exchanger assembly is capable of cooling the exhaust of the third slot 2508C and the fourth slot 2508D, (2) the first heat exchanger assembly is capable of cooling the exhaust of the first slot 2508A, the second slot 2508B, and the third slot 2508C, while the second heat exchange assembly is capable of cooling the exhaust of the fourth slot 2508D, etc.).

[0175] Fig.26 Yes Fig.18 1800 and heat exchangers 1814, 1816, and includes an example cooling system controller circuit 2602. Fig.26 In the example shown, the liquid coolant distribution system 2600 includes an example inlet coolant manifold 2604, an example outlet coolant manifold 2606, an example pump 2608, and an example sensor 2610. Fig.26 In the example shown, the example inlet coolant manifold 2604 couples the example main inlet line 2612, the example chassis inlet line 2614, the example first heat exchanger inlet line 2616, and the example second heat exchanger inlet line 2618. Fig.26 In the example shown, the example inlet coolant manifold 2604 includes an example first valve 2620 and an example second valve 2622. Fig.26 In the example shown, the example outlet coolant manifold 2606 is coupled to the example main outlet line 2624, the example chassis outlet line 2626, the example first heat exchanger outlet line 2628, and the example first heat exchanger outlet line 2630. Fig.26 In the example shown, the example outlet coolant manifold 2606 includes an example third valve 2632 and an example fourth valve 2634. Although reference is made to Fig.18 Chassis 1800 and Fig.18 The heat exchangers 1814, 1816 of the embodiment of the present invention describe the liquid coolant distribution system 2600, but the cooling system controller circuit 2602, manifolds 2604, 2606, pump 2608 and sensor 2610 can be used with other systems (including Fig.25The chassis 2500 and the heat exchanger assembly 2502 are used in combination. Fig.26 In the example shown, the manifolds 2604 , 2606 include an example fifth valve 2636 and an example sixth valve 2638 that regulate coolant flow from the chassis 1800 .

[0176] The inlet coolant manifold 2604 is a mechanical structure that distributes fresh liquid coolant (e.g., cooled liquid coolant, new liquid coolant, etc.) between the cooling system of the chassis 1800 and the internal flow paths of the first heat exchanger 1814 and the second heat exchanger 1816. Fig.26 In the example shown, the inlet coolant manifold 2604 includes valves 2620, 2622, 2636 that control the flow rate of liquid coolant through lines 2616, 2618, 2614, respectively. Fig.26 In the example shown, inlet coolant manifold 2604 is a separate structure from chassis 1800. In other examples, inlet coolant manifold 2604 may be coupled to chassis 1800, to a server rack associated with chassis 1800, to a CDU associated with a server rack, etc. In some examples, inlet coolant manifold 2604 may not be present. In some such examples, the functionality of inlet coolant manifold 2604 may be accomplished by one or more valves of the CDU.

[0177] The outlet coolant manifold 2606 is a mechanical structure that returns stale liquid coolant (e.g., heated liquid coolant, old liquid coolant, etc.) from the cooling system of the chassis 1800 and the internal flow paths of the first heat exchanger 1814 and the second heat exchanger 1816 to the CDU. Fig.26 In the example shown, the outlet coolant manifold 2606 includes valves 2632, 2634, 2638 that control the flow rate of liquid coolant through lines 2626, 2628, 2630, respectively. Fig.26 In the example shown, the inlet coolant manifold 2604 is a separate structure from the chassis 1800. In other examples, the inlet coolant manifold 2604 can be coupled to the chassis 1800, to a server rack associated with the chassis 1800, to a CDU associated with the server rack, etc. Additionally or alternatively, the outlet coolant manifold 2606 can be integral with and / or coupled to the inlet coolant manifold 2604. In some examples, the outlet coolant manifold 2606 may not be present. In some such examples, the functionality of the outlet coolant manifold 2606 can be implemented by one or more valves of the CDU.

[0178] Pump 2608 pumps or drives the liquid coolant through the liquid coolant distribution system 2600. In some examples, the cooling system controller circuit 2602 can control the power output of the pump 2608 to regulate the coolant flow through the liquid coolant distribution system 2600. Fig.26 In the example shown, pump 2608 is external to chassis 1800 and heat exchangers 1814, 1816. In other examples, chassis 1800 and / or one or both of heat exchangers 1814, 1816 may include an internal pump. In some such examples, pump 2608 may be powered by a power source (e.g., Fig. 20 In other examples, pump 2608 may be powered by an external power supply assembly 2000, etc. In some examples, pump 2608 is implemented by a centrifugal pump. In other examples, pump 2608 may be implemented by one or more other suitable types of pumps (e.g., a positive displacement pump, an axial flow pump, a pulse pump, a rotor dynamic pump, etc.) or a combination thereof. In some examples, liquid coolant distribution system 2600 may include additional pumps.

[0179] Sensor 2610 measures and outputs a signal corresponding to liquid coolant distribution system 2600. For example, sensor 2610 may include a temperature sensor that measures the temperature of one or more computing units of chassis 1800 (e.g., Fig.18 of one or more of the computing components 1804, etc.), and the power supply of the chassis 1800 (e.g., Fig. 20 The temperature of one or more power supplies associated with the power supply assembly 2000 (such as the power supply assembly 2000 of the power supply assembly 2000, etc.), the temperature of the exhaust air flow entering the heat exchanger 1814, 1816, the temperature of the coolant in one or more of the coolant lines 2614, 2616, 2618, 2626, 2628, 2630, etc. In some examples, the sensor 2610 may include one or more thermocouples, one or more resistance temperature detectors, one or more thermistors, one or more infrared optical sensors, one or more semiconductor-based sensors. In some examples, the sensor 2610 may include one or more flow meters that measure the flow rate of the coolant through one or more of the lines 2614, 2616, 2618, 2626, 2628, 2630. In some examples, the sensor 2610 may include one or more thermal mass flow sensors, one or more Coriolis flow sensors, one or more differential pressure flow sensors, one or more magnetic flow meters, one or more multiphase flow meters, one or more ultrasonic flow meters, and / or one or more vortex flow meters. Additionally or alternatively, sensor 2610 may have one or more pressure sensors, one or more fluid level sensors, and / or any other suitable type of sensor.

[0180] The lines 2614, 2616, 2618, 2626, 2628, 2630 are conduits (e.g., tubes, pipes, etc.) that contain liquid coolant and move the liquid coolant between the various components of the liquid coolant distribution system 2600. In some examples, some or all of the lines 2614, 2616, 2618, 2626, 2628, 2630 may be implemented by flexible tubes (e.g., rubber tubes, plastic tubes, etc.). Additionally or alternatively, one or two of the lines 2614, 2616, 2618, 2626, 2628, 2630 may be rigid or substantially rigid tubes (e.g., metal pipes, plastic tubes, etc.). In some examples, the lines 2614, 2616, 2618, 2626, 2628, 2630 may be insulated to reduce heat transfer between the liquid coolant and the surrounding environment of the liquid coolant distribution system 2600. In some examples, lines 2614, 2616, 2618, 2626, 2628, 2630 may include sealing features (e.g., seals, gaskets, etc.) to prevent leakage of liquid coolant in lines 2614, 2616, 2618, 2626, 2632, 2634. Fig.26 In the example shown, the lines 2614, 2626 are coupled to the inlet 1810 and the outlet 1812, respectively. In some examples, the coupling between the lines 2614, 2626 and the inlet 1810 and the outlet 1812 may include a quick disconnect (QD) connector. In some such examples, the inlet 1810 and / or the outlet 1812 include a switchable self-locking mechanism that can be enabled / disabled to enable the coupling of the lines 2614, 2626, respectively. In other examples, the lines 2614, 2626 may be coupled to the inlet 1810 and the outlet 1812, respectively. In some examples, the lines 2614, 2626 and / or the valves 2636, 2638 may not be present (e.g., the chassis 1800 does not include a liquid coolant system, etc.).

[0181] The valves 2620, 2622, 2632, 2634, 2636, 2638 regulate (e.g., throttling, etc.) the flow rate and pressure of the liquid coolant through the lines 2614, 2616, 2618, 2626, 2628, 2630, respectively. In some examples, the valves 2620, 2622, 2632, 2634 can be used to regulate the flow rate of the coolant based on the cooling requirements of the power supply (e.g., cooling requirements, heat output, etc.) and / or the computing components associated with the heat exchangers 1814, 1816. In some examples, the valves 2636, 2638 control the flow rate of the liquid coolant through the cooling system associated with the chassis 1800. For example, the position of the first valve 2620 and / or the third valve 2632 can be used to control the flow rate of the liquid coolant through the first heat exchanger 1814. For example, the position of the second valve 2622 and / or the fourth valve 2634 may be used to control the flow rate of the liquid coolant through the second heat exchanger 1816 .

[0182] The cooling system controller circuit 2602 is communicatively coupled to the pump 2608, the sensor 2610, and the valves 2620, 2622, 2632, 2634, 2636, 2638. In some examples, the cooling system controller circuit 2602 can control the position of the valves 2620, 2622, 2632, 2634 based on one or more outputs of the sensor 2610. For example, the cooling system controller circuit 2602 can estimate the cooling demand based on the temperature of the power supply associated with the heat exchanger 1814, 1816, based on the power output of the power supply associated with the heat exchanger 1814, 1806, based on the power input of the power supply associated with the heat exchanger 1814, 1806, based on the temperature and / or flow rate of the exhaust air entering the heat exchanger 1814, 1816. In some examples, the cooling system controller circuit 2602 can determine the cooling instruction based on the determined one or more cooling demands. In some examples, cooling system controller circuit 2602 can base cooling instructions on information related to the workload on computing component 1804. In some examples, cooling system controller circuit 2602 can control the position of one or more of valves 2620, 2622, 2632, 2634, 2636, 2638 and / or the power input of pump 2608 based on the determined cooling instructions. In some examples, cooling system controller circuit 2602 can cause heat exchangers 1814, 1816 to exhaust air at a temperature approximately equal to the ambient temperature. Fig. 27 Cooling system controller circuit 2602 is described in additional detail.

[0183] Fig. 27 Is used for operation Fig.26 Liquid Coolant Distribution System 2600 Fig.26A block diagram of an example implementation of a cooling system controller circuit 2602 is shown. Fig. 27 In the example shown, cooling system controller circuit 2602 includes example interface circuit 2702 , example cooling requirement determiner circuit 2704 , example instruction generator circuit 2706 , and example system controller circuit 2708 . Fig. 27 The cooling system controller circuit 2602 may be instantiated (e.g., created as an instance, generated for any length of time, embodied, realized, etc.) by a programmable circuit such as a central processing unit (CPU) executing a first instruction. Additionally or alternatively, Fig. 27 The cooling system controller circuit 2602 may be instantiated (e.g., created as an instance, produced in any length of time, embodied, implemented, etc.) by (i) an application specific integrated circuit (ASIC) and / or (ii) a field programmable gate array (FPGA), which is structured and / or configured to perform operations corresponding to the first instruction in response to executing the second instruction. Therefore, it should be understood that Fig. 27 Some or all of the circuits may be instantiated at the same or different times. Fig. 27 Some or all of the circuits of can be instantiated, for example, in one or more threads that execute simultaneously on hardware and / or serially on hardware. Furthermore, in some examples, Fig. 27 Some or all of the circuits may be implemented by microprocessor circuits executing instructions and / or FPGA circuits performing operations to implement one or more virtual machines and / or containers.

[0184] The interface circuit 2702 accesses, requests, and / or receives information from the chassis (e.g., Fig.18 and 26The interface circuit 2702 may receive sensor data from sensors associated with the chassis 1800, the chassis 1800, etc.) and / or the heat exchangers 1814, 1816. For example, the interface circuit 2702 may receive sensor data from the sensor 2610. In some examples, the interface circuit 2702 may receive temperature sensor data (e.g., one or more temperatures of one or more computing components 1804 of the chassis 1800, one or more temperatures of one or more power supplies of the chassis 1800, the ambient temperature of the rack 1800, etc.). In some examples, the interface circuit 2702 may receive flow rate sensor data (e.g., the flow rate of liquid coolant through the cooling system 1806 of the chassis 1800, the flow rate of liquid coolant through one or more of the lines 2614, 2616, 2618, 2626, 2628, 2630, the flow rate of exhaust air from one or more power supplies of the chassis 1800, etc.). Additionally or alternatively, the interface circuit 2702 may receive any other suitable sensor data (e.g., pressure data, etc.). In some examples, interface circuit 2702 may convert data received from sensor 2610 into digital form (eg, human-readable, etc.).

[0185] Additionally or alternatively, interface circuit 2702 may access workload distribution information associated with one or more computing units of chassis 1800. For example, interface circuit 2702 may access information related to the workload on one or more computing units of chassis 1800. For example, interface circuit 2702 may access one or more databases associated with chassis 1800 and / or an operator of chassis 1800 to determine the current and / or upcoming workload on the computing units of chassis 1800. In other examples, interface circuit 2702 may access workload distribution information in any other suitable manner. In some examples, interface circuit 2702 is instantiated by programmable circuitry that executes interface instructions and / or is configured to perform operations such as Fig.28A and 28B The operations represented by the flowchart.

[0186] In some examples, the cooling system controller circuit 2602 includes means for interfacing. For example, the means for interfacing can be implemented by the interface circuit 2702. In some examples, the interface circuit 2702 can be implemented by, for example, Fig.29 The example programmable circuit 2912 of the programmable circuit is instantiated. For example, the interface circuit 2702 can be composed of Fig.30 , which executes machine-executable instructions such as those given by Fig.28A At least blocks 2802, 2804 and Fig.28BIn some examples, the cooling system controller circuit 2602 may be instantiated by a hardware logic circuit that may be implemented by Fig.31 The interface circuit 2702 may be implemented by an ASIC, XPU, or FPGA circuit 3100 that is configured and / or constructed to perform operations corresponding to machine-readable instructions. Additionally or alternatively, the interface circuit 2702 may be instantiated by any other combination of hardware, software, and / or firmware. For example, the interface circuit 2702 may be implemented by at least one or more hardware circuits (e.g., processor circuits, discrete and / or integrated analog and / or digital circuits, FPGAs, ASICs, XPUs, comparators, operational amplifiers (op-amps), logic circuits, etc.) that are configured and / or constructed to execute some or all of the machine-readable instructions and / or to execute some or all of the operations corresponding to the machine-readable instructions without executing software or firmware, but other structures are equally suitable.

[0187] Cooling requirement determiner circuit 2704 determines one or more power supplies (e.g., Fig.23 2000, 2302, etc.). For example, the cooling requirement determiner circuit 2704 can determine the cooling requirement of each of the components associated with the heat exchangers in the heat exchangers 1814, 1816 (e.g., the cooling requirement of the components that exhaust air via the first slot 1808A, the cooling requirement of the components that exhaust air via the second slot 1808B, etc.). In some examples, the cooling requirement determiner circuit 2704 can determine the cooling requirement based on the current temperature of the power supply, the flow rate of the exhaust air through the power supply, the temperature of the exhaust air leaving the power supply, and the ambient temperature received from the sensor 2610. Additionally or alternatively, the cooling requirement determiner circuit 2704 can determine the cooling requirement based on the upcoming and / or current workload on the computing components associated with one or more power supplies (e.g., processors powered by the power supplies upstream of the heat exchangers 1814, 1816, etc.).

[0188] In some examples, the cooling requirement determiner circuit 2704 can determine the cooling requirement as an amount of energy (e.g., heat to be dissipated, etc.), an amount of power (e.g., a rate of heat to be dissipated, etc.), and / or a flow rate value (e.g., a flow rate of a coolant required to cool one or more power supplies). In some examples, the cooling requirement determiner circuit 2704 can determine the cooling requirement based on a lookup table (e.g., cooling requirement as an output of the lookup table, sensor data and / or workload information as an input to the lookup table, etc.). Additionally or alternatively, the cooling requirement determiner circuit 2704 can include and / or be implemented by a machine learning model with sensor data and / or workload information as input and cooling requirement as output. In some examples, the cooling requirement determiner circuit 2704 can determine the cooling requirement of the power supply in any other suitable manner (e.g., based on user input, etc.). In some examples, the cooling requirement determiner circuit 2704 is instantiated by a programmable circuit that executes coolant determiner instructions and / or is configured to execute instructions such as Fig.28A and Fig.28B The operations represented by the flowchart.

[0189] In some examples, the cooling system controller circuit 2602 includes means for analyzing the workload. For example, the means for analyzing the workload can be implemented by the cooling demand determiner circuit 2704. In some examples, the cooling demand determiner circuit 2704 can be implemented by, for example, Fig.29 The example programmable circuit 2912 of the example programmable circuit 2912 is instantiated. For example, the cooling demand determiner circuit 2704 can be composed of Fig.30 , which executes machine-executable instructions such as those given by Fig.28A At least box 2806 and Fig.28B In some examples, the cooling demand determiner circuit 2704 may be instantiated by a hardware logic circuit that may be implemented by Fig.31The cooling demand determiner circuit 2704 may be implemented by an ASIC, XPU, or FPGA circuit 3100 that is configured and / or constructed to perform operations corresponding to machine-readable instructions. Additionally or alternatively, the cooling demand determiner circuit 2704 may be instantiated by any other combination of hardware, software, and / or firmware. For example, the cooling demand determiner circuit 2704 may be implemented by at least one or more hardware circuits (e.g., processor circuits, discrete and / or integrated analog and / or digital circuits, FPGAs, ASICs, XPUs, comparators, operational amplifiers (op-amps), logic circuits, etc.) that are configured and / or constructed to execute some or all of the machine-readable instructions and / or to execute some or all of the operations corresponding to the machine-readable instructions without executing software or firmware, but other structures are equally suitable.

[0190] Instruction generator circuit 2706 determines (e.g., generates, creates, etc.) cooling instructions based on the cooling demand. For example, instruction generator circuit 2706 may determine instructions for pump 2608 and / or one or more of valves 2620, 2622, 2632, 2634, 2636, 2638 based on the determined cooling demand determined during execution of block 2806. In some examples, the one or more cooling instructions determined by instruction generator circuit 2706 may include changing one or more positions of one or more of valves 2620, 2622, 2632, 2634, 2636, 2638, a power output of pump 2608, and / or one or more fans associated with chassis 1800 (e.g., Fig. 20 In other examples, instruction generator circuit 2706 may determine any other suitable instructions. For example, instruction generator circuit 2706 may determine cooling instructions via one or more mathematical calculations, lookup tables, and / or machine learning models. In other examples, instruction generator circuit 2706 may determine cooling instructions in any other suitable manner. In some examples, instruction generator circuit 2706 is instantiated by a programmable circuit that executes instruction generator instructions and / or is configured to execute instructions such as Fig.28A and 28B The operations represented by the flowchart.

[0191] In some examples, the cooling system controller circuit 2602 includes means for generating instructions. For example, the means for generating instructions can be implemented by the instruction generator circuit 2706. In some examples, the instruction generator circuit 2706 can be implemented by, for example, Fig.29 The example programmable circuit 2912 of the example programmable circuit 2912 is instantiated. For example, the instruction generator circuit 2706 can be composed of Fig.30 The example microprocessor 3000 is instantiated by the example microprocessor 3000, which performs operations such as Fig.28A In some examples, instruction generator circuit 2706 may be instantiated by hardware logic circuitry, which may be implemented by Fig.31 The instruction generator circuit 2706 may be implemented by an ASIC, XPU, or FPGA circuit 3100 that is configured and / or constructed to perform operations corresponding to machine-readable instructions. Additionally or alternatively, the instruction generator circuit 2706 may be instantiated by any other combination of hardware, software, and / or firmware. For example, the instruction generator circuit 2706 may be implemented by at least one or more hardware circuits (e.g., processor circuits, discrete and / or integrated analog and / or digital circuits, FPGAs, ASICs, XPUs, comparators, operational amplifiers (op-amps), logic circuits, etc.) that are configured and / or constructed to execute some or all of the machine-readable instructions and / or to execute some or all of the operations corresponding to the machine-readable instructions without executing software or firmware, but other structures are equally suitable.

[0192] The system controller circuit 2708 sends instructions to the pump 2608 and / or the valves 2620, 2622, 2632, 2634, 2636, 2638 based on the cooling instructions. For example, the system controller circuit 2708 can send a signal (e.g., a hydraulic signal, a pneumatic signal, an electronic signal, etc.) to one or more controllable features (e.g., an actuator, etc.) of the valves 2620, 2622, 2632, 2634, 2636, 2638. In other examples, the system controller circuit 2708 can send instructions to control the position of the valves 2620, 2622, 2632, 2634, 2636, 2638 via a direct mechanical connection (e.g., a control arm, etc.). In some examples, the system controller circuit 2708 can control the flow rate of the liquid coolant through the pump 2608 by sending a signal (e.g., a hydraulic signal, a pneumatic signal, an electronic signal, etc.). In some such examples, the system controller circuit 2708 can send instructions to increase and / or decrease the power input and / or output of the pump 2608. In some examples, the system controller circuit 2708 is instantiated by a programmable circuit that executes system controller instructions and / or is configured to perform operations such as Fig.28A and 28B The operations represented by the flowchart.

[0193] In some examples, the cooling system controller circuit 2602 includes means for controlling the cooling system. For example, the means for controlling the cooling system can be implemented by the system controller circuit 2708. In some examples, the system controller circuit 2708 can be implemented by, for example, Fig.29 The example programmable circuit 2912 of the system controller circuit 2708 can be instantiated by Fig.30, which executes machine-executable instructions such as those given by Fig.28A At least frames 2810, 2812 and Fig.28B In some examples, the system controller circuit 2708 may be instantiated by a hardware logic circuit, which may be implemented by blocks 2820, 2826, 2830, and 2834 of FIG. Fig.31 The system controller circuit 2708 may be implemented by an ASIC, XPU, or FPGA circuit 3100 that is configured and / or constructed to perform operations corresponding to machine-readable instructions. Additionally or alternatively, the system controller circuit 2708 may be instantiated by any other combination of hardware, software, and / or firmware. For example, the system controller circuit 2708 may be implemented by at least one or more hardware circuits (e.g., processor circuits, discrete and / or integrated analog and / or digital circuits, FPGAs, ASICs, XPUs, comparators, operational amplifiers (op-amps), logic circuits, etc.) that are configured and / or constructed to execute some or all of the machine-readable instructions and / or to execute some or all of the operations corresponding to the machine-readable instructions without executing software or firmware, but other structures are equally suitable.

[0194] Although in Fig. 27 The implementation is shown in Fig.26 The cooling system controller circuit 2602 is an example of a circuit 2602, but Fig. 27 One or more of the elements, processes and / or devices shown in the drawings may be combined, split, rearranged, omitted, eliminated and / or implemented in any other manner. In addition, the interface circuit 2702, the cooling demand determiner circuit 2704, the instruction generator circuit 2706, the system controller circuit 2708 and / or more generally, Fig. 27 The example cooling system controller circuit 2602 may be implemented by hardware alone or by hardware in combination with software and / or firmware. Thus, for example, the interface circuit 2702, the cooling demand determiner circuit 2704, the instruction generator circuit 2706, the system controller circuit 2708, and / or more generally, any of the example cooling system controller circuit 2602 may be implemented by programmable circuitry in combination with machine-readable instructions (e.g., firmware or software), a processor circuit, one or more analog circuits, one or more digital circuits, one or more logic circuits, one or more programmable processors, one or more programmable microcontrollers, one or more graphics processing units (GPUs), one or more digital signal processors (DSPs), one or more ASICs, one or more programmable logic devices (PLDs), and / or one or more field programmable logic devices (FPLDs) such as FPGAs. Furthermore, in addition to Fig. 27In addition to or in place of the elements, processes, and / or devices shown in Fig. 27 The elements, processes and / or devices shown in Fig. 27 The example cooling system controller circuit 2602 may also include one or more elements, processes and / or devices, and / or may also include more than one of any or all of the illustrated elements, processes and / or devices.

[0195] Represents a programmable circuit that can be executed to implement and / or instantiate Fig. 27 The example machine readable instructions and / or representations of the cooling system controller circuit 2602 may be executed by a programmable circuit to implement and / or instantiate Fig. 27 A flow chart of an example operation of the refrigeration system controller circuit 2602 is shown in FIG. Fig.28A and 28B The machine-readable instructions may be one or more executable programs or one or more portions of one or more executable programs executed by a programmable circuit, such as the following in combination with Fig.29 The programmable circuit 2912 shown in the example programmable circuit platform 2900 discussed above and / or may be a combination of the following Fig.30 In some examples, machine-readable instructions enable operations, tasks, etc. to be implemented and / or performed in an automated manner in the real world. As used herein, "automation" refers to situations where there is no human involvement.

[0196] The program may be embodied in instructions (e.g., software and / or firmware) stored on one or more non-transitory computer-readable and / or machine-readable storage media, such as cache memory, magnetic storage devices or disks (e.g., floppy disks, hard disk drives (HDDs), etc.), optical storage devices or optical disks (e.g., Blu-ray disks, compact disks (CDs), digital versatile disks (DVDs), etc.), redundant arrays of independent disks (RAIDs), registers, ROMs, solid-state drives (SSDs), SSD memories, non-volatile memories (e.g., electrically erasable programmable read-only memories (EEPROMs), flash memory, etc.), volatile memories (e.g., any type of random access memory (RAM), etc.), and / or any other storage devices or storage disks. The instructions of the non-transitory computer-readable and / or machine-readable media may be programmed and / or executed by programmable circuits located in one or more hardware devices, but the entire program and / or portions thereof may alternatively be executed and / or instantiated and / or embodied in dedicated hardware by one or more hardware devices other than programmable circuits. The machine-readable instructions may be distributed across multiple hardware devices and / or executed by two or more hardware devices (e.g., a server and a client hardware device). For example, a client hardware device may be implemented by an endpoint client hardware device (e.g., a hardware device associated with a human and / or machine user) or an intermediate client hardware device gateway (e.g., a radio access network (RAN)) that may facilitate communications between a server and an endpoint client hardware device. Similarly, a non-transitory computer-readable storage medium may include one or more media. Furthermore, although reference is made to Fig.28A and Fig.28BThe flowchart shown in describes an example program, but alternatively, many other methods of implementing the example cooling system controller circuit 2602 may be used. For example, the execution order of the blocks of the flowchart may be changed, and / or some of the blocks described may be changed, eliminated, or combined. Additionally or alternatively, any or all of the blocks of the flowchart may be implemented by one or more hardware circuits (e.g., processor circuits, discrete and / or integrated analog and / or digital circuits, FPGAs, ASICs, comparators, operational amplifiers (op-amps), logic circuits, etc.) that are configured to perform corresponding operations without executing software or firmware. The programmable circuits may be distributed in different network locations and / or locally in one or more hardware devices (e.g., single-core processors (e.g., single-core CPUs), multi-core processors (e.g., multi-core CPUs, XPUs, etc.)). For example, the programmable circuit may be a CPU and / or FPGA located in the same package (e.g., the same integrated circuit (IC) package or in two or more separate housings), one or more sub-processors in a single machine, multiple processors distributed on multiple servers in a server rack, multiple processors distributed on one or more server racks, etc., and / or any one or more combinations thereof.

[0197] The machine-readable instructions described herein may be stored in one or more of a compressed format, an encrypted format, a segmented format, a compiled format, an executable format, a packaged format, etc. The machine-readable instructions as described herein may be stored as data (e.g., computer-readable data, machine-readable data, one or more bits (e.g., one or more computer-readable bits, one or more machine-readable bits, etc.), a bitstream (e.g., a computer-readable bitstream, a machine-readable bitstream, etc.), etc.) or a data structure that can be used to create, manufacture and / or generate machine-executable instructions (e.g., stored as one or more portions of instructions, code, a representation of code, etc.). For example, the machine-readable instructions may be segmented and stored on one or more storage devices, disks, and / or computing devices (e.g., servers) located in the same or different locations of a network or a collection of networks (e.g., in the cloud, in an edge device, etc.). The machine-readable instructions may require one or more of installation, modification, adaptation, update, combination, supplementation, configuration, decryption, decompression, unpacking, distribution, reassignment, compilation, etc., so that they are directly readable, interpretable, and / or executable by a computing device and / or other machine. For example, machine-readable instructions may be stored in multiple parts that are separately compressed, encrypted, and / or stored on separate computing devices, where the parts, when decrypted, decompressed, and / or combined, form a set of computer-executable and / or machine-executable instructions that implement one or more functions and / or operations that may together form a program such as described herein.

[0198] In another example, the machine-readable instructions may be stored in a state where they can be read by programmable circuits but require the addition of a library (e.g., a dynamic link library (DLL)), a software development kit (SDK), an application programming interface (API), etc., in order to execute the machine-readable instructions on a particular computing device or other device. In another example, the machine-readable instructions may need to be configured (e.g., stored settings, data inputs, recorded network addresses, etc.) before the machine-readable instructions and / or corresponding one or more programs can be executed in whole or in part. Therefore, as used herein, machine-readable, computer-readable and / or machine-readable media may include instructions and / or one or more programs, regardless of the specific format or state of the machine-readable instructions and / or one or more programs.

[0199] The machine-readable instructions described herein may be represented in any past, present or future instruction language, scripting language, programming language, etc. For example, the machine-readable instructions may be represented in any of the following languages: C, C++, Java, C#, Perl, Python, JavaScript, Hypertext Markup Language (HTML), Structured Query Language (SQL), Swift, etc.

[0200] As mentioned above, Fig.28A and Fig.28BThe example operations of may be implemented using executable instructions (e.g., computer-readable and / or machine-readable instructions) stored on one or more non-transitory computer-readable and / or machine-readable media. As used herein, the terms non-transitory computer-readable medium, non-transitory computer-readable storage medium, non-transitory machine-readable medium, and / or non-transitory machine-readable storage medium are expressly defined to include any type of computer-readable storage device and / or storage disk, and exclude propagating signals and exclude transmission media. Examples of such non-transitory computer-readable medium, non-transitory computer-readable storage medium, non-transitory machine-readable medium, and / or non-transitory machine-readable storage medium include optical storage devices, magnetic storage devices, HDDs, flash memory, read-only memory (ROM), CDs, DVDs, caches, any type of RAM, registers, and / or any other storage device or storage disk, wherein information is stored in the storage device for any duration (e.g., permanently for an extended period of time, for a transient instance, for temporary buffering and / or for caching information). As used herein, the terms "non-transitory computer-readable storage device" and "non-transitory machine-readable storage device" are defined to include any physical (mechanical, magnetic and / or electrical) hardware to retain information over a period of time, but exclude propagating signals and exclude transmission media. Examples of non-transitory computer-readable storage devices and / or non-transitory machine-readable storage devices include any type of random access memory, any type of read-only memory, solid-state memory, flash memory, optical disks, magnetic disks, disk drives, and / or redundant arrays of independent disks (RAID) systems. As used herein, the term "device" refers to a physical structure such as a mechanical and / or electrical device, hardware and / or circuit that may or may not be configured and / or manufactured to execute computer-readable instructions, machine-readable instructions, etc.

[0201] As used herein, "include" and "comprising" (and all forms and tenses thereof) are open-ended terms. Thus, whenever a claim employs any form of "include" or "comprising" (e.g., includes, comprises, includes, including, having, etc.) as a preamble or in any kind of claim statement, it should be understood that other elements, terms, etc. may be present without exceeding the scope of the corresponding claim or statement. As used herein, when the phrase "at least" is used as a transitional term, such as in the preamble of a claim, it is open-ended in the same manner as the terms "include" and "comprising" are open-ended. When the term "and / or" is used, such as in a form such as A, B, and / or C, it refers to any combination or subset of A, B, and C, such as (1) only A, (2) only B, (3) only C, (4) A and B, (5) A and C, (6) B and C, or (7) A and B and C. As used herein in the context of describing structures, components, items, objects, and / or things, the phrase “at least one of A and B” is intended to refer to embodiments that include: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, items, objects, and / or things, the phrase “at least one of A or B” is intended to refer to embodiments that include: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. As used herein in the context of describing the performance or execution of processes, instructions, actions, activities, and / or steps, the phrase “at least one of A and B” is intended to refer to embodiments that include: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing the performance or execution of a process, instruction, action, activity, and / or step, the phrase "at least one of A or B" is intended to refer to embodiments that include: (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.

[0202] As used herein, singular references (e.g., "a," "an," "first," "second," etc.) do not exclude a plurality. As used herein, the term "a" or "an" object refers to one or more of the object. The terms "a" (or "an"), "one or more," and "at least one" are used interchangeably herein. In addition, although listed separately, multiple devices, elements, or actions may be implemented by, for example, the same entity or object. In addition, although individual features may be included in different examples or claims, these features may be combined, and inclusion in different examples or claims does not mean that the combination of features is not feasible and / or advantageous.

[0203] Fig.28Ais executable, instantiated and / or executed by a programmable circuit to operate Fig.26 2600 and / or a flow chart of example operations 2800 for a liquid coolant distribution system. Fig.28A The example machine readable instructions and / or example operations 2800 begin at block 2802 where the interface circuit 2702 accesses a CPU associated with a chassis (e.g., Fig.18 and 26 For example, the interface circuit 2702 may receive sensor data associated with the chassis 1800, the chassis 1800, etc. For example, the interface circuit 2702 may receive sensor data from the sensor 2610. In some examples, the interface circuit 2702 may receive temperature sensor data (e.g., one or more temperatures of one or more computing components 1804 of the chassis 1800, one or more temperatures of one or more power supplies of the chassis 1800, the ambient temperature of the rack 1800, etc.). In some examples, the interface circuit 2702 may receive flow rate sensor data (e.g., the flow rate of liquid coolant through the cooling system 1806 of the chassis 1800, the flow rate of liquid coolant through one or more of the lines 2614, 2616, 2618, 2626, 2628, 2630, the flow rate of exhaust air from one or more power supplies of the chassis 1800, etc.). Additionally or alternatively, the interface circuit 2702 may receive any other suitable sensor data (e.g., pressure data, etc.). In some examples, interface circuit 2702 may convert data received from sensor 2610 into digital form (eg, human-readable, etc.).

[0204] At block 2804, interface circuitry 2702 accesses workload distribution information. For example, interface circuitry 2702 may access information related to workload on one or more computing units of chassis 1800. For example, interface circuitry 2702 may access one or more databases associated with chassis 1800 and / or an operator of chassis 1800 to determine current and / or upcoming workloads on computing units of chassis 1800. In other examples, interface circuitry 2702 may access workload distribution information in any other suitable manner. In some examples, block 2804 is omitted, and control of the system is based on sensor data without regard to workload information.

[0205] At block 2806, the cooling requirement determiner circuit 2704 determines a cooling requirement on one or more power supplies of the chassis based on the workload distribution information and / or the sensor data. For example, the cooling requirement determiner circuit 2704 may determine a cooling requirement for each of the components associated with a heat exchanger in the heat exchangers 1814, 1816 (e.g., a cooling requirement for a component that exhausts air via the first slot 1808A, a cooling requirement for a component that exhausts air via the second slot 1808B, etc.). In some examples, the cooling requirement determiner circuit 2704 may determine the cooling requirement based on a current temperature of the power supply, a flow rate of exhaust air through the power supply, a temperature of the exhaust air leaving the power supply, an ambient temperature received from the sensor 2610. In some examples, the cooling requirement determiner circuit 2704 may determine the cooling requirement based on enabling the heat exchangers 1814, 1816 to exhaust air at a temperature substantially equal to the ambient temperature.

[0206] Additionally or alternatively, the cooling demand determiner circuit 2704 can determine the cooling demand based on the upcoming and / or current workload on the computing components associated with one or more power supplies (e.g., a processor powered by the power supply upstream of the heat exchanger 1814, 1816, etc.). In some examples, the cooling demand determiner circuit 2704 can determine the cooling demand as an amount of energy (e.g., heat to be dissipated, etc.), an amount of power (e.g., a rate of heat to be dissipated, etc.) and / or a flow rate value (e.g., a flow rate of a coolant required to cool one or more power supplies). In some examples, the cooling demand determiner circuit 2704 can determine the cooling demand based on a lookup table (e.g., a cooling demand as an output of the lookup table, sensor data and / or workload information as an input to the lookup table, etc.). Additionally or alternatively, the cooling demand determiner circuit 2704 can use a machine learning model with sensor data and / or workload information as input and cooling demand as output. In some examples, the cooling demand determiner circuit 2704 can determine the cooling demand of the power supply in any other suitable manner (e.g., based on user input, etc.).

[0207] At block 2808, instruction generator circuit 2706 determines cooling instructions based on the cooling demand. For example, instruction generator circuit 2706 may determine instructions for pump 2608 and / or one or more of valves 2620, 2622, 2632, 2634, 2636, 2638 based on the determined cooling demand determined during execution of block 2806. In some examples, the one or more cooling instructions determined by instruction generator circuit 2706 may include changing the position of one or more of valves 2620, 2622, 2632, 2634, 2636, 2638, the power output of pump 2608, and / or one or more fans associated with chassis 1800 (e.g., Fig. 20 Instruction generator circuit 2706 may determine any other suitable instructions. For example, instruction generator circuit 2706 may determine the cooling instructions via one or more mathematical calculations, lookup tables, and / or machine learning models. In other examples, instruction generator circuit 2706 may determine the cooling instructions in any other suitable manner.

[0208] At block 2810, the system controller circuit 2708 sends instructions to the pump 2608 and / or valves 2620, 2622, 2632, 2634, 2636, 2638 based on the cooling instructions. For example, the system controller circuit 2708 can send a signal (e.g., a hydraulic signal, a pneumatic signal, an electronic signal, etc.) to one or more controllable features (e.g., an actuator, etc.) of the valves 2620, 2622, 2632, 2634, 2636, 2638. In other examples, the system controller circuit 2708 can send instructions to control the position of the valves 2620, 2622, 2632, 2634, 2636, 2638 via a direct mechanical connection (e.g., a control arm, etc.). In some examples, the system controller circuit 2708 can control the flow rate of the pump 2608 by sending a signal (e.g., a hydraulic signal, a pneumatic signal, an electronic signal, etc.). In some such examples, system controller circuit 2708 may send instructions to increase and / or decrease power input and / or output of pump 2608 .

[0209] At block 2812, the system controller circuit 2708 determines whether to update the coolant distribution. For example, the system controller circuit 2708 may determine to update the coolant distribution based on changes in the output of the sensor 2610 detected by the interface circuit 2702 and / or changes in the workload on the computing component 1804 of the chassis 1800. Additionally or alternatively, the system controller circuit 2708 may determine to update the coolant distribution based on user settings and / or user commands. Additionally or alternatively, the system controller circuit 2708 may determine to update the coolant distribution periodically (e.g., every minute, every hour, every day, etc.). If the system controller circuit 2708 determines to update the coolant distribution, the operation 2800 returns to block 2802. If the system controller circuit 2708 determines not to update the coolant distribution, the operation 2800 ends.

[0210] Fig.28B is a flow chart representative of example machine readable instructions and / or example operations 2814 that may be executed, instantiated, and / or performed by a programmable circuit to operate a first heat exchanger 1814 . Fig.28BThe example machine readable instructions and / or example operations 2814 begin at block 2816 where the interface circuit 2702 accesses sensor data associated with the rack 1800 and / or the first heat exchanger 1814. For example, the interface circuit 2702 may access sensor data associated with the rack 1800 and / or the first heat exchanger 1814. Fig.24 The interface circuit 2702 may receive sensor data from one or more of the sensors 2402, 2404, 2406, 2408. Additionally or alternatively, the interface circuit 2702 may receive any other suitable sensor data (e.g., pressure data, etc.). In some examples, the interface circuit 2702 may convert the data received from the sensors 2402, 2404, 2406, 2408 into a digital form (e.g., human readable, etc.).

[0211] At block 2818, the cooling demand determiner circuit 2704 determines whether the inlet temperature of the first heat exchanger 1814 is higher than the ambient temperature and lower than the first threshold temperature. For example, the cooling demand determiner circuit 2704 may determine whether the temperature output by the first sensor 2402 is higher than the ambient temperature of the first heat exchanger 1814 (e.g., the temperature of the data center including the chassis 1800, etc.) and lower than the first threshold temperature. In some examples, the first threshold temperature may be based on a user setting, a type of coolant used by the first heat exchanger 1814, one or more hardware components of the chassis 1800, and / or any other suitable metric. In some examples, the first threshold temperature may be 50 degrees Celsius. Additionally or alternatively, the first threshold temperature may be based on the temperature of municipal water that may be used to cool the liquid coolant. If the cooling demand determiner circuit 2704 determines that the inlet temperature of the first heat exchanger 1814 is higher than the ambient temperature and lower than the first threshold temperature, operation 2814 proceeds to block 2818. If the cooling demand determiner circuit 2704 determines that the inlet temperature of the first heat exchanger 1814 is not above the ambient temperature and below the first threshold temperature, then operation 2814 proceeds to block 2824 .

[0212] At block 2820, the system controller circuit 2708 sets the system (e.g., system 2600, etc.) to have a first coolant flow rate and / or sets the coolant inlet temperature to ambient temperature. For example, the system controller circuit 2708 may provide a first coolant flow rate to the system (e.g., system 2600, etc.) by supplying a first coolant flow rate to the pump associated with the first heat exchanger 1814 (e.g., Fig.26 2608, etc.) to change the flow rate of the coolant flow through the first heat exchanger 1814. For example, the system controller circuit 2708 may send a command to a CDU associated with the first heat exchanger 1814 (e.g., Fig.26The system controller circuit 2708 may send instructions to change the temperature of the coolant flowing through the first heat exchanger 1814 to the ambient temperature. In some examples, the system controller circuit 2708 may cause the first heat exchanger 1814 to operate at the minimum coolant flow rate available to the chassis 1800 to reduce power usage of the first heat exchanger 1814, etc. In some examples, the system controller circuit 2708 monitors the temperature of the coolant flowing through the first heat exchanger 1814 via feedback from the third sensor 2406 and / or the fourth sensor 2408.

[0213] At block 2822, the interface circuit 2702 determines whether the outlet temperature of the first heat exchanger 1814 is lower than the inlet temperature of the first heat exchanger 1814. For example, the interface circuit 2702 determines whether the first heat exchanger 1814 is capable of cooling the air flow from the chassis 1800 at the currently set coolant flow rate and coolant temperature by comparing the inlet temperature of the temperature output by the first sensor 2402 with the temperature output by the second sensor 2404. Additionally or alternatively, the interface circuit 2702 may determine whether the first heat exchanger 1814 is reducing the temperature of the air flow flowing therethrough at the currently set coolant flow rate and coolant temperature. If the interface circuit 2702 determines that the outlet temperature of the first heat exchanger 1814 is lower than the inlet temperature of the first heat exchanger 1814, operation 2814 proceeds to block 2834. If the interface circuit 2702 determines that the outlet temperature of the first heat exchanger 1814 is not lower than the inlet temperature of the first heat exchanger 1814, operation 2814 returns to block 2818.

[0214] At block 2824, the cooling demand determiner circuit 2704 determines whether the inlet temperature of the first heat exchanger 1814 is above the first threshold temperature and below the second threshold temperature. For example, the cooling demand determiner circuit 2704 may determine whether the temperature output by the first sensor 2402 is above the first threshold temperature and below the second threshold temperature that is higher than the first threshold temperature. In some examples, the second threshold temperature may be based on a user setting, a type of coolant used by the first heat exchanger 1814, one or more hardware components of the chassis 1800, the first threshold temperature, and / or any other suitable metric. In some examples, the second threshold temperature may be 70 degrees Celsius. If the cooling demand determiner circuit 2704 determines that the inlet temperature of the first heat exchanger 1814 is above the first threshold temperature and below the second threshold temperature, operation 2814 proceeds to block 2826. If the cooling demand determiner circuit 2704 determines that the inlet temperature of the first heat exchanger 1814 is not above the first threshold temperature and below the second threshold temperature, operation 2814 proceeds to block 2830.

[0215] At block 2826, the system controller circuit 2708 sets the system (e.g., system 2600, etc.) to have a second coolant flow rate and / or sets the coolant inlet temperature to be below the ambient temperature. For example, the system controller circuit 2708 may provide a second coolant flow rate to the system (e.g., system 2600, etc.) by supplying a second coolant flow rate to the pump associated with the first heat exchanger 1814 (e.g., Fig.26 2608, etc.) to increase the coolant flow rate through the first heat exchanger 1814 to a second flow rate that is higher than the flow rate associated with block 2820. For example, the system controller circuit 2708 may increase the coolant flow rate through the first heat exchanger 1814 by sending an instruction to a CDU associated with the first heat exchanger 1814 (e.g., Fig.26 The system controller circuit 2708 may send instructions to change the temperature of the coolant flowing through the first heat exchanger 1814 to be lower than the ambient temperature. In some examples, the system controller circuit 2708 may monitor the temperature of the coolant flowing through the first heat exchanger 1814 via feedback from the third sensor and / or the fourth sensor 2408.

[0216] At block 2828, the interface circuit 2702 determines whether the outlet temperature of the first heat exchanger 1814 is lower than the inlet temperature of the first heat exchanger 1814. For example, the interface circuit 2702 determines whether the first heat exchanger 1814 is capable of cooling the air flow from the chassis 1800 at the currently set coolant flow rate and coolant temperature by comparing the inlet temperature of the temperature output by the first sensor 2402 with the temperature output by the second sensor 2404. Additionally or alternatively, the interface circuit 2702 may determine whether the first heat exchanger 1814 is reducing the temperature of the flow flowing therethrough at the currently set coolant flow rate and coolant temperature. If the interface circuit 2702 determines that the outlet temperature of the first heat exchanger 1814 is lower than the inlet temperature of the first heat exchanger 1814, the operation 2814 proceeds to block 2834. If the interface circuit 2702 determines that the outlet temperature of the first heat exchanger 1814 is not lower than the inlet temperature of the first heat exchanger 1814, the operation 2814 returns to block 2824.

[0217] At block 2829, the cooling demand determiner circuit 2704 determines whether the inlet temperature of the first heat exchanger 1814 is above the second threshold temperature. For example, the cooling demand determiner circuit 2704 may determine whether the temperature output by the first sensor 2402 is above the second threshold temperature. If the cooling demand determiner circuit 2704 determines that the inlet temperature of the first heat exchanger 1814 is above the second threshold temperature, then operation 2814 proceeds to block 2830. If the cooling demand determiner circuit 2704 determines that the inlet temperature of the first heat exchanger 1814 is not above the second threshold temperature (e.g., the inlet temperature of the first heat exchanger 1814 is below the second threshold temperature of the first heat exchanger 1814, the first threshold temperature, and the ambient temperature, etc.), then operation 2814 proceeds to block 2830.

[0218] At block 2830, the system controller circuit 2708 sets the system (e.g., system 2600, etc.) to have a third coolant flow rate and / or sets the coolant inlet temperature to a lower ambient temperature. For example, the system controller circuit 2708 may provide a third coolant flow rate to the system (e.g., system 2600, etc.) by supplying a third coolant flow rate to the pump associated with the first heat exchanger 1814 (e.g., Fig.26 2608, etc.) to increase the coolant flow rate through the first heat exchanger 1814 to a third flow rate that is higher than the flow rates associated with blocks 2820 and 2826. In some examples, the system controller circuit 2708 can set the coolant flow rate to a maximum flow rate that the system 2600 can achieve. For example, the system controller circuit 2708 can increase the coolant flow rate through the first heat exchanger 1814 by sending a command to a CDU associated with the first heat exchanger 1814 (e.g., Fig.26 The system controller circuit 2708 may send instructions (such as the third sensor 2406 and the fourth sensor 2408) to change the temperature of the coolant flowing through the first heat exchanger 1814 to an ambient temperature that is lower than the temperature associated with block 2826. In some examples, the system controller circuit 2708 may set the coolant temperature to the lowest temperature that the system 2600 is capable of achieving. In some examples, the system controller circuit 2708 may monitor the temperature of the coolant flowing through the first heat exchanger 1814 via feedback from the third sensor 2406 and / or the fourth sensor 2408.

[0219] At block 2832, the interface circuit 2702 determines whether the outlet temperature of the first heat exchanger 1814 is lower than the inlet temperature of the first heat exchanger 1814. For example, the interface circuit 2702 determines whether the first heat exchanger 1814 is capable of cooling the air flow from the chassis 1800 at the currently set coolant flow rate and coolant temperature by comparing the inlet temperature of the temperature output by the first sensor 2402 with the temperature output by the second sensor 2404. Additionally or alternatively, the interface circuit 2702 may determine whether the first heat exchanger 1814 is reducing the temperature of the flow flowing therethrough at the currently set coolant flow rate and coolant temperature. If the interface circuit 2702 determines that the outlet temperature of the first heat exchanger 1814 is lower than the inlet temperature of the first heat exchanger 1814, operation 2814 proceeds to block 2834. If the interface circuit 2702 determines that the outlet temperature of the first heat exchanger 1814 is not lower than the inlet temperature of the first heat exchanger 1814, operation 2814 returns to block 2824.

[0220] At block 2834, the system controller circuit 2708 determines whether to update the coolant temperature and / or coolant flow rate. For example, the system controller circuit 2708 may determine to update the coolant distribution based on changes in the outputs of the sensors 2402, 2404, 2406, 2408 detected by the interface circuit 2702. Additionally or alternatively, the system controller circuit 2708 may determine to update the coolant distribution based on user settings and / or user commands. Additionally or alternatively, the system controller circuit 2708 may determine to update the coolant distribution periodically (e.g., every minute, every hour, every day, etc.). If the system controller circuit 2708 determines to update the coolant distribution, operation 2814 returns to block 2816. If the system controller circuit 2708 determines not to update the coolant distribution, operation 2814 ends.

[0221] Fig.29 is a block diagram of an example programmable circuit platform 2900 that is configured to execute and / or instantiate Fig.28A and 28B Example machine readable instructions and / or example operations to implement Fig. 27 The programmable circuit platform 2900 can be, for example, a server, a personal computer, a workstation, a self-learning machine (e.g., a neural network), a mobile device (e.g., a cellular phone, a smart phone, an iPad, etc.). TM tablet computer), personal digital assistant (PDA), Internet home appliance, DVD player, CD player, digital video recorder, Blu-ray player, game console, personal video recorder, set-top box, headset (e.g., augmented reality (AR) headset, virtual reality (VR) headset, etc.) or other wearable device, or any other type of computing and / or electronic device.

[0222] The programmable circuit platform 2900 of the illustrated example includes a programmable circuit 2912. The programmable circuit 2912 of the illustrated example is hardware. For example, the programmable circuit 2912 can be implemented by one or more integrated circuits, logic circuits, FPGAs, microprocessors, CPUs, GPUs, DSPs, and / or microcontrollers from any desired series or manufacturer. The programmable circuit 2912 can be implemented by one or more semiconductor-based (e.g., silicon-based) devices. In this example, the programmable circuit 2912 implements the interface circuit 2702, the cooling demand determiner circuit 2704, the instruction generator circuit 2706, and the system controller circuit 2708.

[0223] The programmable circuit 2912 of the illustrated example includes a local memory 2913 (e.g., cache, registers, etc.). The programmable circuit 2912 of the illustrated example communicates with a main memory 2914, 2916 including a volatile memory 2914 and a non-volatile memory 2916 via a bus 2918. The volatile memory 2914 may be composed of a synchronous dynamic random access memory (SDRAM), a dynamic random access memory (DRAM), Dynamic Random Access Memory The non-volatile memory 2916 may be implemented by flash memory and / or any other desired type of memory device. Access to the main memory 2914, 2916 of the illustrated example is controlled by a memory controller 2917. In some examples, the memory controller 2917 may be implemented by one or more integrated circuits, logic circuits, microcontrollers from any desired family or manufacturer, or any other type of circuit to manage the flow of data to and from the main memory 2914, 2916.

[0224] The programmable circuit platform 2900 of the illustrated example also includes an interface circuit 2920. The interface circuit 2920 can be implemented by hardware according to any type of interface standard, such as an Ethernet interface, a universal serial bus (USB) interface, The interface is implemented by a peripheral component interconnect (PCI) interface, a near field communication (NFC) interface, a peripheral component interconnect (PCI) interface, and / or a peripheral component interconnect express (PCIe) interface.

[0225] In the example shown, one or more input devices 2922 are connected to the interface circuit 2920. The one or more input devices 2922 allow a user (e.g., a human user, a machine user, etc.) to input data and / or commands into the programmable circuit 2912. The one or more input devices 2922 may be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, buttons, a mouse, a touch screen, a trackpad, a trackball, an isopoint device, and / or a voice recognition system.

[0226] One or more output devices 2924 are also connected to the interface circuit 2920 of the illustrated example. One or more output devices 2924 can be implemented, for example, by a display device (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display (LCD), a cathode ray tube (CRT) display, an in-place switch (IPS) display, a touch screen, etc.), a tactile output device, a printer, and / or a speaker. Therefore, the interface circuit 2920 of the illustrated example typically includes a graphics driver card, a graphics driver chip, and / or a graphics processor circuit such as a GPU.

[0227] The interface circuitry 2920 of the illustrated example also includes communication devices, such as transmitters, receivers, transceivers, modems, residential gateways, wireless access points, and / or network interfaces to facilitate data exchange with external machines (e.g., any kind of computing device) over the network 2926. Communications may be accomplished through, for example, an Ethernet connection, a digital subscriber line (DSL) connection, a telephone line connection, a coaxial cable system, a satellite system, a line-of-sight wireless system, a line-of-sight wireless system, a cellular telephone system, an optical connection, and the like.

[0228] The programmable circuit platform 2900 of the illustrated example also includes one or more mass storage disks or devices 2928 for storing firmware, software and / or data. Examples of such mass storage disks or devices 2928 include magnetic storage devices (e.g., floppy disks, drives, HDDs, etc.), optical storage devices (e.g., Blu-ray disks, CDs, DVDs, etc.), RAID systems, and / or solid-state storage disks or devices such as flash memory devices and / or SSDs.

[0229] Can be Fig.28A and Fig.28B The machine-readable commands 2932 implemented by the machine-readable instructions may be stored in the mass storage device 2928, in the volatile memory 2914, in the non-volatile memory 2916, and / or on at least one non-transitory computer-readable storage medium, such as a CD or DVD, which may be removable.

[0230] Fig.30 yes Fig.29 2912 of the programmable circuit 2912. In this example, Fig.29 The programmable circuit 2912 is implemented by the microprocessor 3000. For example, the microprocessor 3000 may be a general-purpose microprocessor (eg, a general-purpose microprocessor circuit). The microprocessor 3000 executes Fig.28A and 28B Some or all of the machine readable instructions of a flowchart to effectively Fig. 27 The circuits are instantiated as logic circuits to perform operations corresponding to those machine-readable instructions. In some such examples, Fig. 27The circuit is instantiated by the hardware circuit of the microprocessor 3000 in combination with machine-readable instructions. For example, the microprocessor 3000 can be implemented by a multi-core hardware circuit, such as a CPU, DSP, GPU, XPU, etc. Although it can include any number of example cores 3002 (for example, 1 core), the microprocessor 3000 of this example is a multi-core semiconductor device including N cores. The cores 3002 of the microprocessor 3000 can operate independently or can cooperate to execute machine-readable instructions. For example, the machine code corresponding to a firmware program, an embedded software program, or a software program can be executed by one of the cores 3002, or can be executed by multiple cores 3002 at the same or different times. In some examples, the machine code corresponding to the firmware program, the embedded software program, or the software program is divided into threads and executed in parallel by two or more cores 3002. The software program may correspond to Fig.28A and 28B A flowchart may represent a portion or all of machine-readable instructions and / or operations.

[0231] The kernel 3002 can communicate via a first example bus 3004. In some examples, the first bus 3004 can be implemented by a communication bus to implement communications associated with one (or more) of the kernels 3002. For example, the bus 3004 can be implemented by at least one of an inter-integrated circuit (I2C) bus, a serial peripheral interface (SPI) bus, a PCI bus, or a PCIe bus. Additionally or alternatively, the bus 3004 can be implemented by any other type of computing or electrical bus. The kernel 3002 can obtain data, instructions, and / or signals from one or more external devices via an example interface circuit 3006. The kernel 3002 can output data, instructions, and / or signals to one or more external devices via the interface circuit 3006. Although the core 3002 of this example includes an example local memory 3020 (e.g., a level 1 (L1) cache that can be divided into an L1 data cache and an L1 instruction cache), the microprocessor 3000 also includes an example shared memory 3010 that can be shared by the cores (e.g., a level 2 (L2 cache)) for high-speed access to data and / or instructions. Data and / or instructions can be transferred (e.g., shared) by writing to and / or reading from the shared memory 3010. The local memory 3020 of each of the core 3002 and the shared memory 3010 can be a cache memory including multiple levels and a main memory (e.g., Fig.29 2914, 2916). Typically, higher level memory in the hierarchy exhibits lower access times and smaller storage capacities than lower level memory. Changes to various levels of the cache hierarchy are managed (e.g., coordinated) by a cache coherence policy.

[0232] Each core 3002 may be referred to as a CPU, DSP, GPU, etc., or any other type of hardware circuit. Each core 3002 includes a control unit circuit 3014, an arithmetic and logic (AL) circuit 3016 (sometimes referred to as an ALU), a plurality of registers 3018, a local memory 3020, and a second example bus 3022. Other structures may exist. For example, each core 3002 may include a vector unit circuit, a single instruction multiple data (SIMD) unit circuit, a load / store unit (LSU) circuit, a branch / jump unit circuit, a floating point unit (FPU) circuit, etc. The control unit circuit 3014 includes a semiconductor-based circuit that is configured to control (e.g., coordinate) data movement within the corresponding core 3002. The AL circuit 3016 includes a semiconductor-based circuit that is configured to perform one or more mathematical and / or logical operations on data within the corresponding core 3002. The AL circuit 3016 of some examples performs integer-based operations. In other examples, the AL circuit 3016 also performs floating-point operations. In other examples, AL circuit 3016 may include a first AL circuit that performs integer-based operations and a second AL circuit that performs floating-point operations. In some examples, AL circuit 3016 may be referred to as an arithmetic logic unit (ALU).

[0233] Registers 3018 are semiconductor-based structures for storing data and / or instructions, such as the results of one or more operations performed by the AL circuit 3016 of the corresponding core 3002. For example, registers 3018 may include one or more vector registers, one or more SIMD registers, one or more general registers, one or more flag registers, one or more segment registers, one or more machine-specific registers, one or more instruction pointer registers, one or more control registers, one or more debug registers, one or more memory management registers, one or more machine check registers, etc. Registers 3018 may be in the form of Fig.30 Alternatively, registers 3018 may be organized in any other arrangement, format, or structure, such as by being distributed throughout core 3002 to shorten access time. Second bus 3022 may be implemented by at least one of an I2C bus, an SPI bus, a PCI bus, or a PCIe bus.

[0234] Each core 3002 and / or more generally, the microprocessor 3000 may include additional and / or alternative structures to those shown and described above. For example, there may be one or more clock circuits, one or more power supplies, one or more power gates, one or more cache home agents (CHAs), one or more convergence / common grid stoppers (CMSs), one or more shifters (e.g., one or more barrel shifters), and / or other circuits. The microprocessor 3000 is a semiconductor device fabricated to include many transistors that are interconnected to implement the above structures in one or more integrated circuits (ICs) contained in one or more packages.

[0235] The microprocessor 3000 may include and / or collaborate with one or more accelerators (e.g., acceleration circuits, hardware accelerators, etc.). In some examples, the accelerator is implemented by a logic circuit to perform certain tasks faster and / or more efficiently than a general-purpose processor. Examples of accelerators include ASICs and FPGAs, such as those discussed herein. GPUs, DSPs, and / or other programmable devices may also be accelerators. The accelerator may be onboard the microprocessor 3000, in the same chip package as the microprocessor 3000, and / or in one or more packages separated from the microprocessor 3000.

[0236] Fig.31 yes Fig.29 3100. In this example, the programmable circuit 2912 is implemented by an FPGA circuit 3100. For example, the FPGA circuit 3100 can be implemented by an FPGA. The FPGA circuit 3100 can be used, for example, to perform operations that could otherwise be performed by Fig.30 However, once configured, the FPGA circuit 3100 instantiates the operations and / or functions corresponding to the machine-readable instructions in hardware, and therefore, can generally perform these operations / functions faster than those performed by a general-purpose microprocessor executing corresponding software.

[0237] More specifically, as described above Fig.30 The microprocessor 3000 (which is a general-purpose device that can be programmed to perform the Fig.28A and Fig.28B In contrast, a flowchart may represent some or all of the machine-readable instructions, but the interconnections and logic circuits are fixed once manufactured. Fig.31 The example FPGA circuit 3100 includes a circuit that can be configured, constructed, programmed, and / or interconnected in different ways after manufacture to instantiate, for example, Fig.28A and Fig.28BThe flowchart represents the interconnections and logic circuits corresponding to some or all of the operations / functions of the machine-readable instructions. In particular, the FPGA circuit 3100 can be thought of as an array of logic gates, interconnections, and switches. The switches can be programmed to change how the logic gates are interconnected through the interconnections, effectively forming one or more dedicated logic circuits (unless and until the FPGA circuit 3100 is reprogrammed). The configured logic circuits enable the logic gates to cooperate in different ways to perform different operations on the data received by the input circuits. These operations can correspond to the operations performed by Fig.28A and Fig.28B In this way, FPGA circuit 3100 can be configured and / or constructed to effectively instantiate the same as Fig.28A and 28B Some or all of the operations / functions corresponding to the machine-readable instructions of the flowchart of the FPGA circuit 3100 can be used as a dedicated logic circuit to perform the operations / functions corresponding to those software instructions in a dedicated manner similar to an ASIC. Fig.28A and Fig.28B Some or all of the machine-readable instructions may correspond to operations / functions faster than those performed by a general-purpose microprocessor.

[0238] exist Fig.31 In some examples, the FPGA circuit 3100 is configured and / or structured in response to being programmed (and / or reprogrammed one or more times) based on a binary file. In some examples, a binary file can be compiled and / or generated based on instructions of a hardware description language (HDL) such as Lucid, Very High Speed ​​Integrated Circuit (VHSIC) Hardware Description Language (VHDL), or Verilog. For example, a user (e.g., a human user, a machine user, etc.) can write code or a program corresponding to one or more operations / functions in an HDL; the code / program can be translated into a low-level language as needed; and the code / program (e.g., code / program in a low-level language) can be converted (e.g., by a compiler, a software application, etc.) into a binary file. In some examples, Fig.31 The FPGA circuit 3100 can access and / or load binary files to enable Fig.31 The FPGA circuit 3100 is configured and / or constructed to perform one or more operations / functions. For example, a binary file may be obtained by a bit stream (e.g., one or more computer-readable bits, one or more machine-readable bits, etc.), data (e.g., computer-readable data, machine-readable data, etc.), and / or Fig.31 The FPGA circuit 3100 is accessible to machine-readable instructions to enable Fig.31 The FPGA circuit 3100 or one or more portions thereof is configured and / or structured.

[0239] In some examples, the binary file is compiled, generated, converted, and / or otherwise output from a unified software platform for programming the FPGA. For example, the unified software platform can translate a first instruction (e.g., code or program) corresponding to one or more operations / functions in a high-level language (e.g., C, C++, Python, etc.) into a second instruction corresponding to one or more operations / functions in the HDL. In some such examples, the binary file is compiled, generated, and / or otherwise output from the unified software platform based on the second instruction. In some examples, Fig.31 The FPGA circuit 3100 can access and / or load binary files to enable Fig.31 The FPGA circuit 3100 is configured and / or constructed to perform one or more operations / functions. For example, a binary file may be obtained by a bit stream (e.g., one or more computer-readable bits, one or more machine-readable bits, etc.), data (e.g., computer-readable data, machine-readable data, etc.), and / or Fig.31 The FPGA circuit 3100 is accessible to machine-readable instructions to enable Fig.31 The FPGA circuit 3100 or one or more portions thereof is configured and / or structured.

[0240] Fig.31 The FPGA circuit 3100 includes an example input / output (I / O) circuit 3102 to obtain data from and / or output data to an example configuration circuit 3104 and / or external hardware 3106. For example, the configuration circuit 3104 can be implemented by an interface circuit that can obtain a binary file, which can be implemented by a bitstream, data, and / or machine-readable instructions to configure the FPGA circuit 3100 or one or more portions thereof. In some such examples, the configuration circuit 3104 can obtain the binary file from a user, a machine (e.g., a hardware circuit (e.g., a programmable or dedicated circuit) that can implement an artificial intelligence / machine learning (AI / ML) model to generate a binary file, etc.), and / or any one or more combinations thereof. In some examples, the external hardware 3106 can be implemented by an external hardware circuit. For example, the external hardware 3106 can be implemented by Fig.30 It is implemented by the microprocessor 3000.

[0241] FPGA circuit 3100 also includes an array of example logic gate circuits 3108, a plurality of example configurable interconnects 3110, and example storage circuits 3112. Logic gate circuits 3108 and configurable interconnects 3110 may be configured to instantiate a logic gate circuit that may correspond to Fig.28A and 28B One or more operations / functions of at least some of the machine-readable instructions and / or other desired operations. Fig.31 The logic gate circuit 3108 shown in is manufactured in the form of blocks or groups. Each block includes a semiconductor-based electrical structure that can be constructed as a logic circuit. In some examples, the electrical structure includes a logic gate (e.g., an AND gate, an OR gate, a NOT gate, etc.), which provides a basic building block for the logic circuit. An electrically controllable switch (e.g., a transistor) is present in each of the logic gate circuits 3108 so that the electrical structure and / or the logic gate can be configured to form a circuit to perform the desired operation / function. The logic gate circuit 3108 may include other electrical structures such as a lookup table (LUT), a register (e.g., a flip-flop or a latch), a multiplexer, etc.

[0242] The configurable interconnect 3110 of the illustrated example is a conductive path, trace, via, or the like, which may include an electrically controllable switch (e.g., a transistor) whose state may be changed by programming (e.g., using an HDL instruction language) to activate or deactivate one or more connections between one or more of the logic gate circuits 3108, thereby programming the desired logic circuit.

[0243] The storage circuit 3112 of the illustrated example is configured to store one or more results of one or more operations performed by the corresponding logic gates. The storage circuit 3112 may be implemented by a register, etc. In the illustrated example, the storage circuit 3112 is distributed between the logic gate circuits 3108 to facilitate access and increase execution speed.

[0244] Fig.31 The example FPGA circuit 3100 also includes an example dedicated operation circuit 3114. In this example, the dedicated operation circuit 3114 includes a dedicated circuit 3116, which can be called to implement common functions to avoid the need to program these functions on site. Examples of such dedicated circuits 3116 include memory (e.g., DRAM) controller circuits, PCIe controller circuits, clock circuits, transceiver circuits, memories, and multiplier accumulator circuits. There may be other types of dedicated circuits. In some examples, the FPGA circuit 3100 may also include an example general-purpose programmable circuit 3118, such as an example CPU 3120 and / or an example DSP 3122. Other general-purpose programmable circuits 3118 may additionally or alternatively exist, such as GPUs, XPUs, etc., which can be programmed to perform other operations.

[0245] although Fig.30 and Fig.31 Shows Fig.29 These are two example implementations of programmable circuit 2912, but many other approaches are contemplated. For example, an FPGA circuit may include an onboard CPU, such as Fig.30 One or more of the example CPUs 3120. Thus, Fig.29The programmable circuit 2912 can be formed by combining at least Fig.30 An example microprocessor 3000 and Fig.31 In some such hybrid examples, Fig.30 One or more kernels 3002 may execute Fig.28A and 28B The flowchart represents a first portion of machine-readable instructions to perform a first one or more operations / one or more functions, Fig.31 The FPGA circuit 3100 may be configured and / or constructed to perform as Fig.28A and 28B The second one or more operations / one or more functions corresponding to the second portion of the machine-readable instructions represented by the flowchart, and / or the ASIC can be configured and / or constructed to perform the same as the ... Fig.28A and 28B The flowchart represents a third portion of machine-readable instructions corresponding to a third one or more operations / one or more functions.

[0246] Therefore, it should be understood that Fig. 27 Some or all of the circuits may be instantiated at the same or different times. Fig.30 The same and / or different one or more portions of the microprocessor 3000 may be programmed to execute one or more portions of the machine-readable instructions at the same and / or different times. In some examples, Fig.31 The same and / or different one or more portions of the FPGA circuit 3100 may be configured and / or constructed to perform operations / functions corresponding to one or more portions of the machine-readable instructions at the same and / or different times.

[0247] In some examples, Fig. 27 Some or all of the circuits of may be instantiated, for example, in one or more threads of execution in parallel and / or serially. Fig.30 The microprocessor 3000 can execute machine-readable instructions in one or more threads of execution in parallel and / or serially. In some examples, Fig.31 The FPGA circuit 3100 can be configured and / or constructed to perform operations / functions in parallel and / or serially. In addition, in some examples, Fig. 27 Some or all of the circuitry may be in Fig.30 The microprocessor 3000 is implemented within one or more virtual machines and / or containers.

[0248] In some examples, Fig.29 The programmable circuit 2912 can be in one or more packages. For example, Fig.30 The microprocessor 3000 and / or Fig.31 The FPGA circuit 3100 can be in one or more packages. In some examples, the XPU can be composed of Fig.29 The XPU may be implemented in a programmable circuit 2912, which may be in one or more packages. For example, the XPU may include a CPU in a package (e.g., Fig.30 Microprocessor 3000, Fig.31 CPU 3120, etc.), a DSP in another package (e.g., Fig.31 DSP 3122 of ), a GPU in yet another package, and an FPGA in yet another package (e.g., Fig.31 FPGA circuit 3100).

[0249] Example methods, devices, systems, and articles for compact aftermarket modular heat exchangers are disclosed herein. Other examples and combinations thereof include the following:

[0250] Example 1 includes an apparatus including a heat exchanger having an air flow inlet and an air flow outlet, a rack frame for receiving components of a server chassis, and a bracket coupled to the rack frame for holding the heat exchanger near a rear of the server chassis.

[0251] Example 2 includes the apparatus of Example 1, wherein the heat exchanger includes a body defining an interior, the air flow inlet for receiving exhaust flow from the server chassis, the air flow outlet defining a first flow path through the interior with the air flow inlet, and a tube at least partially disposed within the interior, the tube defining a second flow path between the air flow inlet and the air flow outlet, the tube being used to receive a liquid coolant.

[0252] Example 3 includes the apparatus of Example 1, further comprising a duct directing exhaust flow from the server chassis to the heat exchanger, the duct comprising a duct inlet defining a first plane and a duct outlet defining a second plane, the second plane having a smaller area than the first plane.

[0253] Example 4 includes the apparatus of Example 3, wherein the bracket is to be disposed adjacent a rear slot of the server chassis, the duct inlet being adjacent to the server chassis via coupling of the bracket to the heat exchanger and the server chassis.

[0254] Example 5 includes the apparatus of Example 3, wherein the duct includes a first plate extending from the duct outlet away from the duct inlet, the first plate being coupled to the heat exchanger, and a second plate extending from the duct outlet away from the duct inlet, the second plate being parallel to the first plate, the second plate being coupled to the heat exchanger.

[0255] Example 6 includes the apparatus of Example 5, wherein the conduit further comprises a flow conduit comprising a first planar sidewall and a second sidewall comprising a corner.

[0256] Example 7 includes the apparatus of Example 6, wherein the first plate, the second plate, and the flow conduit are an integral component.

[0257] Example 8 includes the apparatus of Example 5, wherein the heat exchanger is disposed between the first plate and the second plate.

[0258] Example 9 includes the apparatus of Example 1, wherein the component is a power supply component of the server chassis.

[0259] Example 10 includes the apparatus of Example 1, wherein the bracket frame and the bracket are integral, the bracket frame is disposed within the server chassis, and the bracket protrudes from the server chassis.

[0260] Example 11 includes a server chassis comprising a frame sized to be supported in a server rack, a computing unit disposed within the frame, a power supply for supplying power to the computing unit, the power supply including an exhaust outlet, and a heat exchanger supported by the frame, the heat exchanger being disposed downstream of a fluid passage associated with the exhaust outlet, the heat exchanger including an internal flow path for receiving a liquid coolant.

[0261] Example 12 includes the server chassis of Example 11, wherein the exhaust outlet is a first exhaust outlet, the power supply is a first power supply, the heat exchanger is a first heat exchanger, the fluid pathway is a first fluid pathway, and the server chassis further includes a second power supply, a second exhaust outlet associated with the second power supply, and a second heat exchanger, wherein the second heat exchanger is disposed downstream of the second fluid pathway associated with the second exhaust outlet.

[0262] Example 13 includes the server chassis of Example 12, wherein the internal flow path is a first internal flow path, the second heat exchanger includes a second internal flow path, and the first internal flow path and the second internal flow path are configured to receive coolant from a coolant distribution unit associated with the server chassis.

[0263] Example 14 includes the server chassis of Example 11, wherein the exhaust outlet is a first exhaust outlet, the fluid pathway is a first fluid pathway, the power supply is a first power supply, and the server chassis further includes a second power supply and a second exhaust outlet associated with the second power supply, and the heat exchanger is disposed downstream of the second fluid pathway associated with the second exhaust outlet.

[0264] Example 15 includes the server chassis of Example 11, further comprising a liquid cooling system for cooling the computing unit, the liquid cooling system fluidly coupled to a manifold, the manifold fluidly coupled to the internal flow path of the heat exchanger.

[0265] Example 16 includes the server chassis of Example 11, wherein the frame defines an interior of the server chassis, the power supply is to be disposed within the interior, the computing unit is to be disposed within the interior, and the heat exchanger is to be outside the interior.

[0266] Example 17 includes a non-transitory machine-readable storage medium comprising instructions for causing a programmable circuit to at least access sensor data from a sensor associated with a server chassis, determine a cooling requirement for a power supply of the server chassis based on the sensor data, and modify a flow rate of a liquid coolant through a heat exchanger supported by the server chassis based on the cooling requirement, the heat exchanger being disposed outside the server chassis and downstream of an exhaust port of the power supply.

[0267] Example 18 includes the non-transitory machine-readable storage medium of Example 17, wherein the power supply is a first power supply, the heat exchanger is a first heat exchanger, the cooling requirement is a first cooling requirement, the flow rate is a first flow rate, the exhaust port is a first exhaust port, and the instructions cause the programmable circuit to determine a second cooling requirement of a second power supply of the server chassis based on the sensor data and to modify a second flow rate of the liquid coolant through a second heat exchanger based on the second cooling requirement, the second heat exchanger being positioned downstream of a second exhaust port of the second power supply.

[0268] Example 19 includes the non-transitory machine-readable storage medium of Example 17, wherein the instructions cause the programmable circuit to modify the flow rate such that the heat exchanger exhausts air at a temperature substantially equal to an ambient temperature of an environment surrounding the server chassis.

[0269] Example 20 includes the non-transitory machine-readable storage medium of Example 17, wherein the instructions cause the programmable circuit to modify the flow rate via at least one of a pump or a valve.

[0270] The attached claims are incorporated into this detailed description by reference. Although certain example systems, devices, articles, and methods have been disclosed herein, the coverage of this patent is not limited thereto. On the contrary, this patent covers all systems, devices, articles, and methods that fully fall within the scope of the claims of this patent.

Claims

1. A device comprising: a heat exchanger having an air flow inlet and an air flow outlet; A bracket frame, the bracket frame is used to receive components of the server chassis; as well as A bracket is coupled to the rack frame, the bracket being used to hold the heat exchanger proximate a rear portion of the server chassis.

2. The device according to claim 1, wherein: The heat exchanger comprises: a body defining an interior, the air flow inlet for receiving exhaust air from the server chassis, the air flow outlet and the air flow inlet defining a first flow path through the interior; and A tube is at least partially disposed within the interior, the tube defining a second flow path between the air flow inlet and the air flow outlet, the tube being adapted to receive a liquid coolant.

3. The apparatus of claim 1 , further comprising a conduit directing exhaust flow from the server chassis to the heat exchanger, the conduit comprising: a conduit inlet defining a first plane; as well as A duct outlet is defined in a second plane having a smaller area than the first plane.

4. The device according to claim 3, wherein: The bracket is to be disposed near a rear slot of the server chassis, and the duct inlet is adjacent to the server chassis via the coupling of the bracket to the heat exchanger and the server chassis.

5. The device according to claim 3, wherein: The pipeline includes: a first plate extending from the duct outlet away from the duct inlet, the first plate coupled to the heat exchanger; and A second plate extends from the duct outlet away from the duct inlet, the second plate being parallel to the first plate, the second plate being coupled to the heat exchanger.

6. The device according to claim 5, wherein: The pipeline also includes a flow conduit, the flow conduit including: a planar first sidewall; and The second side wall includes a corner.

7. The device according to claim 6, wherein: The first plate, the second plate and the flow duct are an integral component.

8. The device according to claim 5, wherein: The heat exchanger is disposed between the first plate and the second plate.

9. The device according to any one of claims 1 to 8, wherein: The component is a power supply assembly of the server chassis.

10. The device according to any one of claims 1 to 8, wherein: The support frame and the bracket are integrated, the support frame is disposed within the server chassis, and the bracket protrudes from the server chassis.

11. A server chassis, comprising: a frame sized so that the frame is supported in a server rack; a computing unit, the computing unit being arranged in the frame; a power supply, the power supply being used to supply power to the computing unit, the power supply comprising an exhaust outlet; as well as A heat exchanger is supported by the frame, the heat exchanger being disposed downstream of the fluid passage associated with the exhaust outlet, the heat exchanger including an internal flow path for receiving a liquid coolant.

12. The server chassis according to claim 11, wherein: The exhaust outlet is a first exhaust outlet, the power supply is a first power supply, the heat exchanger is a first heat exchanger, the fluid pathway is a first fluid pathway, and the server chassis further comprises: Second power supply; a second exhaust outlet associated with the second power source; and A second heat exchanger is disposed downstream of a second fluid passage associated with the second exhaust outlet.

13. The server chassis according to claim 12, wherein: The internal flow path is a first internal flow path, the second heat exchanger includes a second internal flow path, the first internal flow path and the second internal flow path receive coolant from a coolant distribution unit associated with the server chassis.

14. The server chassis according to claim 11, wherein: The exhaust outlet is a first exhaust outlet, the fluid pathway is a first fluid pathway, the power source is a first power source, and the server chassis further comprises: a second power source; and A second exhaust outlet is associated with the second power source, the heat exchanger being disposed downstream of a second fluid passage associated with the second exhaust outlet.

15. The server chassis of any one of claims 11-14, further comprising a liquid cooling system for cooling the computing unit, the liquid cooling system being fluidly coupled to a manifold, the manifold being fluidly coupled to an internal flow path of the heat exchanger.

16. The server chassis according to any one of claims 11 to 14, wherein: The frame defines an interior of the server chassis, the power supply is disposed within the interior, the computing unit is disposed within the interior, and the heat exchanger is located outside the interior.

17. A method comprising: accessing sensor data from sensors associated with the server chassis; determining a cooling requirement for a power supply of the server chassis based on the sensor data; and Based on the cooling demand, a flow rate of liquid coolant through a heat exchanger supported by the server chassis, the heat exchanger being disposed external to the server chassis and downstream of an exhaust port of the power supply, is modified.

18. The method according to claim 17, wherein: The power source is a first power source, the heat exchanger is a first heat exchanger, the cooling demand is a first cooling demand, the flow rate is a first flow rate, the exhaust port is a first exhaust port, and the method further comprises: determining a second cooling requirement for a second power supply of the server chassis based on the sensor data; and Based on the second cooling requirement, a second flow rate of liquid coolant through a second heat exchanger is modified, the second heat exchanger being positioned downstream of a second exhaust outlet of the second power source.

19. The method of claim 17, further comprising modifying the flow rate so that the heat exchanger exhausts air at a temperature substantially equal to an ambient temperature of an environment surrounding the server chassis.

20. The method of claim 17, further comprising modifying the flow rate via at least one of a pump or a valve.

21. A machine-readable storage medium comprising instructions for causing a machine to perform a method as defined in any one of claims 17 to 20.

22. An apparatus comprising means for performing a method as defined in any one of claims 17 to 20.