Floating connector mechanism

Abstract

The present disclosure relates to floating connector mechanisms. A connector connectable between a rack and a switch includes a coupling member and a fixed interface member, the coupling member connected to a piston assembly operably coupled to a body structure, the fixed interface member coupled to the body structure. The piston assembly includes a piston connected to the coupling member, the piston displaceable relative to the body structure in axial and radial directions, thereby enabling the coupling member to move in axial and radial directions. The connector can be contained in a switch.

Description

Technical Field

[0001] This disclosure relates to connector mechanisms, and more specifically, to connector mechanisms for connecting rack-mounted equipment to a rack. Background Technology

[0002] For liquid-cooled rack-mountable devices (e.g., rack switches and blade servers), inserting them into the rack may require "blind mating," meaning all connections between the rack-mounted device and the rack must be aligned without visual confirmation, including any quick-disconnect connections such as Universal Quick Disconnect (UQD) connections. To accommodate tolerances, the connector mechanism on the rack-mounted device must be able to float and compensate for misalignment, a capability currently achieved by connector designs. However, in current connector designs, any movement outside the rack-mounted device via the connector results in a corresponding movement inside the device, which is undesirable because such movement requires all conduits coupled to the connector to be flexible and able to move with the connector during blind mating. Therefore, rigid connections cannot be used on the rack-mounted device side of the connector mechanism. Summary of the Invention

[0003] A connector for connecting between a rack and a switch includes a coupling member and a fixed interface member. The coupling member is connected to a piston assembly operably coupled to a body structure, and the fixed interface member is coupled to the body structure. The piston assembly includes a piston connected to the coupling member, which is axially and radially displaceable relative to the body structure, thereby enabling axial and radial movement of the coupling member while keeping the component connected to the rack or switch stationary for connection. This connector can be included in a switch. This connector can be included in all blind-fit liquid-cooled switches / servers.

[0004] In one embodiment, the connector includes: a fixed interface member coupled to a body structure such that the fixed interface member is fixed relative to the body structure; and a piston assembly operatively coupled to and connected to the fixed interface member via the body structure, the piston assembly including: a piston connected to the coupling member; the piston being located in the piston assembly and configured to be axially and radially displaced relative to the body structure such that the coupling member is movable axially and radially.

[0005] In one embodiment, the connector connecting a rack and a switch includes: a fixed interface member coupled to a body structure such that the fixed interface member is fixed relative to the body structure; the fixed interface member is connected to the switch; and a piston assembly operatively coupled to the body structure and connected to the fixed interface member via the body structure. The piston assembly includes: a piston connected to a coupling member, which is connected to the rack. The piston is located within the piston assembly and configured to be axially and radially displaced relative to the body structure, thereby allowing axial and radial floating on the rack side of the connector while remaining stationary on the switch side of the connector. Therefore, the connector can float and compensate for misalignment to accommodate tolerances.

[0006] In one embodiment, the liquid-cooled switch includes a housing, tubing within the housing for delivering coolant, and a connector with two degrees of freedom configured to counteract the dependence of movement within and outside the switch, such that when a portion of the connector moves outside the switch, the tubing within the switch connected to the connector does not move.

[0007] Other aspects of this disclosure will become apparent from the following detailed description taken in conjunction with the accompanying drawings, which illustrate the principles of this disclosure. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of a computing system that can implement the embodiments of the present disclosure.

[0009] Figure 2 This is a block diagram of a rack-mounted device according to an embodiment of the present disclosure.

[0010] Figure 3 This is a block diagram of a rack-mounted device with an equipment-level cooling system according to an embodiment of the present disclosure.

[0011] Figure 4 This is a cross-sectional perspective view of a connector according to an embodiment of the present disclosure.

[0012] Figure 5 This is a cross-sectional side view of a connector according to an embodiment of the present disclosure.

[0013] Figure 6 This is an exploded view of a connector according to an embodiment of the present disclosure.

[0014] Figure 7A This is a perspective front view of the mounting block of the connector according to an embodiment of the present disclosure.

[0015] Figure 7B This is a rear perspective view of the mounting block of the connector according to an embodiment of the present disclosure.

[0016] Figure 8A This is a front perspective view of the piston hole of a connector according to an embodiment of the present disclosure.

[0017] Figure 8B This is a cross-sectional side view of the piston bore of a connector according to an embodiment of the present disclosure.

[0018] Figure 9A This is a perspective view of the piston of a connector according to an embodiment of the present disclosure.

[0019] Figure 9B This is a cross-sectional side view of the piston of the connector according to an embodiment of the present disclosure.

[0020] Figure 10A This is a perspective view of the closed barb of a connector according to an embodiment of the present disclosure.

[0021] Figure 10B This is another perspective view of the closed barb of the connector according to an embodiment of the present disclosure.

[0022] Figure 11A This is a perspective view of the locking nut of a connector according to an embodiment of the present disclosure.

[0023] Figure 11B This is a cross-sectional side view of the locking nut of the connector according to an embodiment of the present disclosure.

[0024] Figure 12 This is a schematic diagram of a switch having two connectors according to an embodiment of the present disclosure.

[0025] Figure 13 The illustration shows a computer system according to at least one embodiment.

[0026] Figure 14A and Figure 14B The diagram illustrates a network architecture according to an embodiment of the present disclosure.

[0027] Figure 15 The illustration shows an example network configuration of components that can be used to implement various aspects of the embodiments.

[0028] Figure 16 An exemplary data center with a cooling system according to an embodiment of the present disclosure is illustrated.

[0029] Figure 17 The illustration depicts a client-server network comprising a plurality of interconnected network server computers according to an embodiment of the present disclosure.

[0030] Figure 18 The illustration depicts a computer network connecting one or more computing machines according to an embodiment of the present disclosure.

[0031] Figures 19A-19BThe illustrations depict networked computer systems according to different embodiments of the present disclosure.

[0032] Figure 20 The illustration shows a device for enabling neural networks that is directly mounted in a liquid-cooled server rack according to an embodiment of the present disclosure.

[0033] Figure 21A The illustration shows components of a system for data center rack liquid cooling that can be implemented using a neural network-enabled method, according to an embodiment of the present disclosure.

[0034] Figure 21B The illustration depicts a system employing a reverse-return design according to an embodiment of the present disclosure.

[0035] Figure 21C The diagram illustrates a control system according to an embodiment of the present disclosure.

[0036] Figure 21D A system according to an embodiment of the present disclosure is illustrated.

[0037] In this specification, similar reference numerals may be used to identify similar elements. Detailed Implementation

[0038] It should be understood that the components, as generally described herein and in the examples shown in the accompanying drawings, can be arranged and designed in a variety of different configurations. Therefore, the following more detailed descriptions of the various examples shown in the accompanying drawings are not intended to limit the scope of this disclosure, but merely to illustrate representative examples. While various aspects of the examples are given in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise explicitly stated.

[0039] This disclosure may be implemented in other specific forms without departing from its spirit or essential characteristics. The described embodiments should be considered in all respects as illustrative rather than limiting. Therefore, the scope of this disclosure is indicated by the appended claims rather than this detailed description. All modifications falling within the meaning and equivalents of the claims should be included within the scope of the claims.

[0040] The features, advantages, or similar expressions mentioned in this specification do not imply that all features and advantages achievable with this disclosure should or actually exist in any single example. Rather, references to features and advantages should be understood as implying that a particular feature, advantage, or characteristic described in connection with a particular example is included in at least one example of this disclosure. Therefore, the discussion of features and advantages and similar expressions in this specification may (but not necessarily) refer to the same embodiment.

[0041] Furthermore, the features, advantages, and characteristics described herein can be combined in any suitable manner in one or more embodiments. Those skilled in the art will recognize from the description herein that this disclosure can be practiced without having one or more specific features or advantages of a particular embodiment. In other instances, other features and advantages may be recognized in certain examples, and these features and advantages may not be present in all examples of this disclosure.

[0042] The terms "an example," "example," or similar expressions used in this specification refer to a particular feature, structure, or characteristic described in connection with the illustrated example that is included in at least one embodiment of this disclosure. Therefore, phrases such as "in an example," "in the example," and similar expressions appearing in this specification may refer to the same embodiment, but are not necessarily so.

[0043] As mentioned above, liquid-cooled rack-mount equipment (e.g., rack switches and blade servers) may require connectors that can accommodate misalignment when the rack-mount equipment is inserted into the rack. This is because inserting rack-mount equipment into the rack may require "blind fitting," where an operator stands in front of the server rack and pushes the server in to ensure all components are perfectly aligned, regardless of manufacturing tolerances. Traditional connectors are designed to float to adjust for these misalignments during blind fitting. Unfortunately, the design of traditional connectors means that any movement outside the rack-mount equipment will cause a corresponding movement inside the rack-mount equipment through these connectors. This necessitates the use of flexible tubing in the rack-mount equipment, limiting the types of hoses that can be used and posing challenges to the connection process. This movement cannot be separated, causing displacement of internal components during operation. The movement of hoses inside the switch generates friction, leading to problems when the system mates with the rack. This friction can even prevent the hoses from properly connecting to the rack.

[0044] The connector according to the currently disclosed embodiments is designed to allow electrical and fluid connections to float, adjusting for any misalignment during blind mating of rack-mounted equipment to the rack. Furthermore, the connector is designed so that external movement of the rack-mounted equipment does not cause corresponding internal movement through the connector. This decoupling mechanism reduces friction and allows the use of different types of hoses. The connector eliminates friction issues caused by hose movement within the switch, ensuring a smoother connection process. Specifically, the connector can be configured to enable quick disconnection from the rack without interfering with any internal components of the switch or requiring flexible connections within the switch, thus separating rapid disconnection movement external to the switch from movement within the switch and movement between the rack side and the switch / server side. The connector design also allows for the possibility of not using any hoses. As an example, the connector can be directly connected to the manifold of the rack-mounted equipment without any hoses. This connector construction allows for a smaller connector size and enables non-concentric designs.

[0045] Therefore, by using this connector, rigid piping can be used in rack-mounted equipment, thereby increasing the strength of the piping used in rack-mounted equipment and / or reducing the overall cost of rack-mounted equipment.

[0046] Now for reference Figure 1 The figure illustrates a computing system 100 that may implement embodiments of the present disclosure. In some embodiments, the computing system 100 may include a coolant distribution unit 112 capable of supplying coolant to one or more server systems 104A-104C via a coolant supply line 116. The coolant may be directed to one or more servers 108A-108D and / or one or more switches 110 associated with each of the server systems 104A-104C. Used coolant (e.g., coolant heated to cool the various components of the servers 108A-108D, switches 110, and / or server systems 104A-104C) may be returned to the coolant distribution unit 112 via a coolant return line 120, where the used coolant can be cooled, and then recycled back to the server systems 104A-104C via the coolant supply line 116. In one embodiment, one or more of server systems 104A-104C may correspond to a DGX™ A100 branded server rack and / or a DGX™ branded SuperPod™ manufactured by NVIDIA®. Additionally or alternatively, one or more of servers 108A-108D may also correspond to a DGX™ branded server and / or similar product manufactured by NVIDIA®.

[0047] As previously described, each of the server systems 104A-104C may include additional components such as power supplies, storage devices, cables, cooling lines, etc. In some embodiments, each of the server systems 104A-104C may include a frame (referred to herein as a rack) capable of accommodating one or more servers 108A-108D and one or more switches 110. The rack of the server systems 104A-104C may be configured as a cabinet or other enclosure comprising a plurality of rack units (RUs) configured to accommodate equipment associated with the server systems 104A-104C. The servers 108A-108D and switches 110 may be designed to be inserted into or mounted into the rack of the server systems 104A-104C and may occupy one or more RUs. Therefore, the servers 108A-108D and switches 110 may be referred to herein as rack-mounted devices.

[0048] The coolant distribution unit 112 may include at least one pump, a heat exchanger, and a processor capable of controlling the temperature of the coolant directed by the coolant distribution unit 112. In some embodiments, the coolant described herein may correspond to water or any other liquid capable of being cooled by the heat exchanger of the coolant distribution unit 112. In one embodiment, the coolant distribution unit 112 may receive refrigerated liquid (e.g., water) and / or may include a cooler capable of cooling the liquid before it is pumped to the server systems 104A-104C via the coolant supply line 116.

[0049] Now for reference Figure 2 , Figure 2 This is a block diagram of a rack-mounted device 108 according to an embodiment of the present disclosure. Figure 2 In this context, rack-mounted device 108 is a server, which may correspond to at least one of servers 108A-108D in server systems 104A-104C. Server 108 may include a power supply 204, a graphics processing unit (GPU) 208, a power converter 210, a processor 212, memory 214, a fan 216, I / O ports 218, and / or other components 220. In some embodiments, two or more components of server 108 may be interconnected via cables 206, electrical buses and / or communication buses, electrical traces, electrical interconnects, and / or combinations thereof. These interconnections can be used... Figure 2 The lines with arrows in the middle are used to represent the meaning.

[0050] Power supply 204 can provide electrical power to the components of server 108. In some embodiments, power supply 204 may include a transformer and / or other electronic components that prevent overload, regulate power signals, and / or provide backup power to server 108.

[0051] GPU 208 may correspond to one or more computer processing devices, processing units, and / or processing circuitry. For example, GPU 208 may be provided as silicon, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), any other type of integrated circuit (IC) chip, an IC chip assembly, a data processing unit (DPU), a quantum processing unit (QPU), multiple parallel processing units (PPUs), etc. The QPU is configured to perform one or more operations related to quantum algorithms. In one embodiment, each of the one or more QPUs may include multiple qubits, and the one or more QPUs may communicate with each other via a quantum channel. In some embodiments, each of the multiple qubits may include a local qubit, a global qubit, and / or a synchronization qubit. In some embodiments, the local qubit of each QPU may be configured to perform one or more operations related to quantum algorithms on the QPU associated with the local qubit.

[0052] In some examples, the GPU may be provided as a central processing unit (CPU), a microprocessor, or multiple microprocessors. Examples of GPU 208 include, but are not limited to, NVIDIA® GeForce® RTX 2000 series processors, NVIDIA® GeForce® RTX 3000 series processors, AMD Radeon™ RX 5000 series processors, AMD Radeon™ RX 6000 series processors, and / or any other graphics processing unit.

[0053] Power converter 210 may correspond to an electrical circuit or device capable of receiving a first voltage (e.g., a first DC voltage) and generating a different second voltage (e.g., a second DC voltage). For example, power converter 210 can convert an 8V first voltage to a 1V second voltage. Additionally or alternatively, power converter 210 may include multiple components that generate two different voltages from the received first voltage. In this example, power converter 210 can convert a 54V first voltage to a first different voltage of 8V and a second different voltage of 1V. In some embodiments, power converter 210 may include multiple connected power converters.

[0054] Processor 212 may be similar to (if not identical to) the GPU 208 described above, or vice versa. In one embodiment, processor 212 may correspond to one or more computer processing devices, processing units, processing circuits, DSPs, general-purpose microprocessors (e.g., Intel® Core™ i3, i5, i7, or i9 processors; Intel® Celeron® processors; Intel® Xeon® processors; Intel® Pentium® processors; AMD Ryzen™ processors; AMD Athlon™ processors; AMD Phenom™ processors; Apple A10 or 10X Fusion processors; Apple A11, A12, A12X, A12Z, or A13 Bionic processors; or any other general-purpose microprocessor), GPUs (e.g., NVIDIA® GeForce® RTX2000 series processors, NVIDIA® GeForce® RTX3000 series processors, AMD Radeon™ RX5000 series processors, AMD Radeon™ RX6000 series processors, or any other GPU), ASICs, FPGAs, or other equivalent integrated or discrete logic circuit systems.

[0055] In this document, GPU 208 and / or processor 212 may be collectively referred to as a “computing device,” “GPU,” “processing circuitry,” and / or “processing unit.” In some embodiments, GPU 208 and / or processor 212 may correspond to a CPU, GPU, FPGA, ASIC, and / or similar device. Therefore, the term “XPU” may be used to refer to GPU 208 and / or processor 212 and may include any computing architecture and related structures capable of performing at least one processing function of server 108. In some embodiments, XPU may be a multi-purpose programmable device that accepts digital data as input, processes the digital data according to instructions stored in its internal memory, and provides a result as output. Because XPU may have internal memory, XPU may implement sequential digital logic. Like most microprocessors, XPU can operate on numbers and symbols represented in a binary number system.

[0056] Memory 214, or storage memory, can correspond to any type of non-transitory computer-readable medium. In some exemplary embodiments, memory 214 may include volatile or non-volatile memory and a controller for that memory. Non-limiting examples of storage memory 214 that may be used in server 108 may include random access memory (“RAM”), read-only memory (“ROM”), buffer memory, flash memory, solid-state memory, or variations thereof. Any of these memory types can be considered a non-transitory computer memory device, even if the stored data can be changed once or more.

[0057] Fan 216 may correspond to a device capable of moving or directing air across server 108 and / or over one or more component surfaces of server 108. In some embodiments, fan 216 may include one or more blades that, when rotated (e.g., by an electric motor), cause airflow across the components of server 108. In some examples, fan 216 may cause airflow over a heatsink or cooling plate of server 108. The heatsink may be configured to make thermal contact with one or more components of server 108 (e.g., GPU 208 and / or processor 212) that generate heat during operation. When these components generate heat, the heatsink removes heat from the components (e.g., by absorbing heat) and directs at least a portion of the heat to one or more fins or surfaces of the heatsink. Airflow generated by fan 216 may be directed across the fins or surfaces of the heatsink, carrying heat away from the heatsink and dissipating it into the air. In some embodiments, air may be exhausted from server 108. Fresh air may be drawn into server 108 by the operation of fan 216.

[0058] Input / output (I / O) ports 218 may correspond to one or more interconnects, through which server 108 can receive information and / or communicate. Additionally or alternatively, I / O ports 218 may correspond to one or more interconnects, through which server 108 can send information and / or communicate. Examples of I / O ports 218 may include, but are not limited to, Ethernet ports, RJ-11 ports, RJ-45 ports, RS-232 ports, USB ports, and / or combinations thereof.

[0059] Although rack-mounted device 108 has been described as a server, the rack-mounted device described herein can also be a switch or other device that can be inserted into or mounted on a rack. Therefore, the rack-mounted device 108 described herein is not limited to a server.

[0060] Rack-mounted equipment (such as server 108) can be liquid-cooled equipment, which includes equipment-level cooling systems. Figure 3The illustration shows a rack-mounted device 300 with a device-level cooling system 302 according to an embodiment of the present disclosure. In the illustrated embodiment, the rack-mounted device 300 is a server. However, in other embodiments, the rack-mounted device 300 may be a different device, such as a switch that can be inserted into a rack. Figure 3 As shown, the device-level cooling system 302 includes at least one server manifold 304 located within a housing 306, allowing heat transfer fluids (e.g., coolant) to enter and exit the rack-mounted device 300 from the rack. Coolant from the rack manifold enters through an inlet pipe 308 and exits through an outlet pipe 310. Each of the inlet and outlet pipes is connected to a connector 400, which allows the rack-mounted device 300 to be connected to the rack when it is inserted into the rack. A detailed description of the connector 400 follows.

[0061] Inlet pipe 308 and outlet pipe 310 are connected to manifold 304. Coolant from manifold 304 flows from inlet pipe 314 through one or more cold plates 316, via at least one intermediate pipe 318 connecting these cold plates, and returns to manifold 304 through a corresponding outlet pipe 320. This constitutes at least one cooling loop 322 within rack-mounted device 300. In the illustrated embodiment, rack-mounted device 300 includes two cooling loops 322. However, in other embodiments, rack-mounted device 300 may include a single cooling loop or more than two cooling loops. In at least one embodiment, each cold plate 210 is associated with at least one computing component 324, such as a processor or GPU.

[0062] In at least one embodiment, one or more cold plates 316 are configurable cold plates.

[0063] Connector 400 allows the use of various types of hoses, including rigid hoses, flexible hoses, corrugated hoses, or welded hoses, without compensating for internal movement, thus providing greater flexibility in hose selection. In at least one embodiment, multiple intermediate lines, such as intermediate lines 318, may be present in each cooling circuit 322 of the rack-mounted device 300; these lines may be rigid, flexible, corrugated, or welded hoses or conduits. In at least one embodiment, fluid adapters may be provided to couple these lines to the cold plate 316 and / or manifold 304. The dimensions of these fluid adapters may be designed according to the inlet and outlet specifications of the cold plate 316 and manifold 304. The connector also allows for the possibility of not using any hoses; for example, the connector may be directly connected to the manifold 304 or one of the cold plates 316.

[0064] See now Figure 4 and Figure 5The figure shows a connector 400 according to one embodiment of the present disclosure. Figure 4 This is a cross-sectional perspective view of connector 400, and Figure 5 This is a cross-sectional side view of connector 400. Connector 400 is used to connect rack-mounted equipment (such as servers or switches) to a rack, thereby physically coupling the rack-mounted equipment to the rack. Furthermore, connector 400 can also be used to provide a fluid connection between the rack-mounted equipment and the rack, allowing heat transfer fluid to flow between the rack-mounted equipment and the rack, thereby achieving liquid cooling, which refers to the use of a liquid coolant to absorb and dissipate heat from one or more heat-generating components (such as processors and memory modules) in the rack-mounted equipment. To dissipate the absorbed heat more efficiently, the liquid coolant circulates from the rack-mounted equipment to a heat exchanger, such as coolant distribution unit 112. Therefore, connector 400 can be used as a liquid inlet port to supply “cooled” liquid coolant to the rack-mounted equipment; or as a liquid outlet port to transfer “heated” liquid coolant out of the rack-mounted equipment. Thus, in typical applications, a single rack-mounted equipment uses two connectors 400 to transfer liquid coolant in and out of the rack-mounted equipment.

[0065] like Figure 4 and Figure 5 As shown, connector 400 includes: a fixed interface member 426 coupled to a body structure 416 such that the fixed interface member 426 is fixed relative to the body structure 416; and a piston assembly 403 operatively coupled to and operatively coupled to the body structure 416 and the fixed interface member 426. Piston assembly 403 includes a piston 404 connected to coupling member 402. Piston 404 is located in piston assembly 403 and configured to be axially and radially displaced relative to body structure 416, such that coupling member 402 is movable axially and radially. In some embodiments, fixed interface member 426 may include a closing barb, or any fixed interface structure that can be attached to a coolant line of a rack-mounted device. In some embodiments, body structure 416 may be a mounting block, or any structure that can be mounted on a rack-mounted device. In some embodiments, coupling member 402 may be quick-disconnect member 402, or any coupling assembly that can be connected to a corresponding coupling member on the rack. Thus, in this disclosure, coupling member 402 is sometimes referred to as quick-disconnect member, body structure 416 is sometimes referred to as mounting block, and fixed interface member 426 is sometimes referred to as closing barb.

[0066] Quick-disconnect member 402 and closing barb 426 are located at opposite ends of the connector. Quick-disconnect member 402 is designed to mate with a corresponding quick-disconnect member on the rack. As an example, quick-disconnect member 402 may be a universal quick-disconnect (UQD) member. In the illustrated embodiment, quick-disconnect member 402 is a male quick-disconnect member designed to mate with a female quick-disconnect member on the rack. However, in alternative embodiments, quick-disconnect member 402 may be a female quick-disconnect member designed to mate with a male quick-disconnect member on the rack. In other embodiments, quick-disconnect member 402 may be any coupling member capable of connecting to a corresponding coupling member on the rack. Since quick-disconnect member 402 is used for connection to the rack, quick-disconnect member is referred to herein as being located on the rack side of connector 400. As described in more detail below, connector 400 is designed such that quick-disconnect member 402 is a floating member that can move axially and radially (e.g., + / -2 mm), thereby allowing bidirectional floating on the rack side, such as... Figure 5 As shown, this is to compensate for any misalignment when a rack-mounted device with a pair of connectors 400 is inserted into the rack while remaining stationary on the device side (e.g., the switch side).

[0067] The closing barb 426 of connector 400 is used to connect to a coolant line in a rack-mounted device to which the connector is attached. The closing barb 426 can be inserted into the coolant line or conduit to provide fluid connection to a fluid circulation cooling system (e.g., equipment-level cooling system 302) in the rack-mounted device, thereby cooling the various components of the rack-mounted device. In other embodiments, the closing barb 426 can be replaced with any fixed interface member that can be attached to a coolant line in the rack-mounted device. Because this interface member is also attached to mounting block 416, it remains fixed relative to the mounting block even if any axial and / or radial movement occurs in the quick-disconnect member 402. As an example, the closing barb 426 can be replaced with a metal connector that can be welded to a rigid or metal coolant line. Because the closing barb 426 is used to connect to other internal components of the rack-mounted device, the closing barb 426 is referred to herein as being located on the equipment side of connector 400. As described in more detail below, connector 400 is designed such that the closing barb 426 is unaffected by the movement of the quick-disconnect member. In other words, the device side of connector 400 (located inside the rack-mounted device) is unaffected by any movement or motion of the rack side of connector 400 (referring to the outside of the rack-mounted device). Specifically, connector 400 is designed such that even if quick-disconnect member 402 is displaced or moved axially and / or radially, closing barb 426 remains stationary. In other words, connector 400 is designed to counteract the dependence on movement inside and outside the rack-mounted device, such that when a part of the connector (e.g., quick-disconnect member 402) moves outside the rack-mounted device, the piping inside the rack-mounted device connected to the connector will not move. Therefore, rigid piping (e.g., metal piping) can be used to connect to the closing barb 426 of the connector or the device side, which can reduce the overall cost of the rack-mounted device and increase the strength of the piping and its associated connections used in the rack-mounted device.

[0068] When the quick-disconnect member 402 of connector 400 mates with a corresponding quick-disconnect member on the rack, liquid coolant can flow through the connector. If connector 400 serves as an inlet to the rack-mounted equipment to which it is attached, liquid coolant can flow from the rack to the rack-mounted equipment. If connector 400 serves as an outlet to the rack-mounted equipment to which it is attached, coolant can flow from the rack-mounted equipment to the rack. Therefore, by using two such connectors 400, coolant can circulate inside and outside the rack-mounted equipment, thereby regulating the temperature of the various components within the rack-mounted equipment.

[0069] The components of connector 400 will refer to Figure 6 To describe further, Figure 6 This is an exploded view of a connector according to an embodiment of this disclosure. Figure 4-6As shown, connector 400 includes a mounting block 416, which serves as a structural anchor point for other components of the connector. Therefore, the other components of connector 400 are structurally supported by mounting block 416. Thus, mounting block 416 can be considered the body of connector 400. Therefore, mounting block 416 can be referred to herein as the body structure. Mounting block 416 can be used to physically attach connector 400 to rack-mounted devices, such as servers or switches. In the illustrated embodiment, mounting block 416 includes a plurality of holes 417 for screws to mount connector 400 to rack-mounted devices. However, in other embodiments, mounting block 416 may have other types of features that enable mounting connector 400 to rack-mounted devices. Alternatively, mounting block 416 may not include any features for mounting connector to rack-mounted devices. In this case, mounting block 416 can be mounted to rack-mounted devices using external brackets or other methods (e.g., adhesives or solder).

[0070] More detailed illustrations of mounting block 416 are shown below. Figure 7A and Figure 7B middle. Figure 7A This is a perspective front view of mounting block 416, showing the front 720 of the mounting block. Figure 7B This is a perspective rear view of mounting block 416, showing the rear side 722 of the mounting block. As described herein, the front side 720 of mounting block 416 refers to the side facing the rack when the rack-mounted device with connector 400 is being installed into the rack; while the rear side 722 of mounting block 416 refers to the side facing away from the rack when the rack-mounted device with connector 400 is being installed into the rack.

[0071] Mounting block 416 is configured or manufactured to have a central hole 724 extending through the mounting block. Therefore, the front side 720 of mounting block 416 includes a front opening of the central hole 724, while the rear side 722 of mounting block 416 includes a rear opening of the central hole 724. The rear side 722 of mounting block 416 also includes a circular protrusion 726 for connection to a closing barb 426.

[0072] Piston seals 406 and 408 allow axial movement and seal the heat transfer fluid. Piston 404, piston bore assembly 420, and piston seals 406 and 408 enable quick-release member 402 to slide in and out against spring forces. A central bore 724 of mounting block 416 defines an internal region for receiving piston assembly 403, which is connected to quick-release member 402.

[0073] In some embodiments, piston assembly 403 includes anti-friction washers 410 and rigid washers 412 to reduce friction caused by the axial force of the spring. These washers ensure smooth radial movement with minimal force. Piston assembly 403 also includes piston 404, piston seals 406 and 408, a coil spring 414, and piston bore assembly 420. Piston bore assembly 420 includes a tubular portion 419 designed to receive an elongated portion 405 of piston 404.

[0074] The piston bore assembly 420 is located within the connector 400 to prevent axial movement of the piston bore assembly. However, axial movement of the piston 404 is permitted, resulting in a reduction in the overall length of the piston assembly 403 when axial pressure is applied to the piston 404 toward the piston bore assembly 420. This allows the connector 400 to maintain pressure during movement.

[0075] Figure 8A and Figure 8B The piston bore component 420 is shown in more detail. Figure 8A This is a perspective view of the piston bore component 420, showing the circular groove 830 on the back side of the piston bore component. Figure 8B This is a cross-sectional side view of the piston bore assembly 420, showing the inner bore 832 of the tubular portion 419 for receiving the elongated portion 405 of the piston 404.

[0076] Figure 9A and Figure 9B Piston 404 is shown in more detail. Figure 9A This is a perspective view of piston 404, showing circular grooves 940 and 942 on the elongated portion 405 of the piston. Figure 9B This is a cross-sectional side view of piston 404, showing a cavity region 944 in the piston body and a hole 946 penetrating an elongated portion of the piston. The cavity region 944 of piston 404 is used to accommodate a quick-release member 402. The hole 946 of piston 404 is used to allow coolant to flow through the piston.

[0077] Back Figure 5 -7. Quick-release member 402 is attached to piston 404. In the illustrated embodiment, quick-release member 402 is inserted into cavity region 944 of piston 404. Because quick-release member 402 is attached to piston 404, it moves with piston. Thus, any movement of piston 404 will cause the same movement of quick-release member 402.

[0078] A helical spring 414 of piston assembly 403 is located between piston bore member 420 and piston 404, causing the piston with quick-release member 402 to be pushed back to its default position (at which point the piston assembly reaches its maximum length). In one embodiment, piston bore member 420 and piston 404 act as a spring preload mechanism relative to helical spring 414 to ensure that the piston returns to its default position after any axial movement, which corresponds to the maximum length of the piston assembly.

[0079] Piston seals 406 and 408 of piston assembly 403 are located on the elongated portion 405 of piston 404 and serve to ensure a seal between the piston and piston bore assembly 420, thereby supporting axial and radial movement while ensuring leak-free coolant flow. The flow is diverted within the mechanism (rather than the entire block moving). Piston seals 406 and 408 (for axial movement) and end face seal 424 (for radial movement) contribute to this function. These seals provide internal mechanical movement on the frame side while ensuring leak-free flow.

[0080] Specifically, piston seals 406 and 408 are located within circular recesses 940 and 942, respectively, on the elongated portion 405 of piston 404. In some embodiments, piston seals 406 and 408 may be O-rings, which may be made of a polymer (e.g., an elastomer).

[0081] Friction-reducing washers 410 and 412 of piston assembly 403 are located between piston 404 and coil spring 414. Connector 400 also includes another friction-reducing washer 418 located between mounting block 416 and piston bore member 420. For example, friction-reducing washers 410 and 418 may be polytetrafluoroethylene (PTFE) or Teflon washers, while 412 may be a brass washer. Friction-reducing washers 410 and 418 are used to reduce friction when piston assembly 403, having quick-release member 402, moves radially.

[0082] Connector 400 also includes a locking nut 422, an end face seal 424, and a closing barb 426. The closing barb 426 is attached to the mounting block 416. Specifically, the closing barb 426 is secured to a circular protrusion 726 on the mounting block 416. Because the closing barb 426 is attached to the mounting block 416, the closing barb remains stationary relative to the mounting block. When the mounting block 416 is mounted on a rack-mounted device, the mounting block remains stationary relative to the rack-mounted device, which in turn means that the closing barb 426 also remains stationary relative to the rack-mounted device. Therefore, as described above, any conduit directly or indirectly connected to the closing barb 426 need not be flexible, because the conduit does not move relative to the rack-mounted device when the quick-release member 402 of connector 400 moves axially and / or radially.

[0083] Figure 10A and Figure 10B The closed barb 426 is shown in more detail. Figure 10A It is a perspective view of the closed barb 426, showing the inner region 1050 designed to be mounted on the circular protrusion 726 of the mounting block 416. Figure 10B This is another perspective view of the closing barb 426, showing the barb mounting end 1052 of the closing barb, which can be inserted into a pipe. It is noted again here that in other embodiments, the closing barb 426 can be replaced with any interface component that can be attached to a coolant line of a rack-mounted device.

[0084] Back Figure 5 -7. An end face seal 424 is located between the piston bore member 420 and the closing barb 426, serving as a seal between the piston assembly 403 and the closing barb 426. In some embodiments, the end face seal 424 may be an O-ring, which may also be made of a polymer (e.g., an elastomer), similar to piston seals 406 and 408. A lock nut 422 is mounted at the end of the elongated portion 405 of the piston 404 and is used to lock the piston assembly 403 into place and preload the helical spring 414. The lock nut 422 also serves to determine the axial displacement of the piston 404.

[0085] Figure 11A and Figure 11B Lock nut 422 is shown in more detail. Figure 11A This is a perspective view of the locking nut 422, showing the groove 1160. Figure 11B This is a cross-sectional side view of the lock nut 422, showing the groove 1160 and the hole 1162 passing through the lock nut. The groove 1160 on the lock nut 422 is for engaging a tool (not shown), the notch of which can be inserted into the groove to lock the lock nut 422.

[0086] Turn now Figure 12 The figure shows a rack-mounted device 1200 in the form of a switch according to an embodiment of the present disclosure. Figure 12 As shown, rack-mounted device 1200 includes two connectors 400 that mate with corresponding connectors 1204 on rack 1206. In some embodiments, each connector 400 of rack-mounted device 1200 includes a male UQD for connection to a female UQD of rack connector 1204, and vice versa. Rack-mounted device 1200 may include one or more other connectors 1208 that mate with corresponding connectors 1210 on rack. Connectors 1208 may be used for power supply and / or communication.

[0087] During the insertion of rack-mount device 1200 into rack 1206, axial and / or radial misalignment may occur between connector 400 and rack connector 1204 of rack-mount device 1200. For any radial misalignment, the misalignment is corrected by radial movement of piston assembly 403 via friction-reducing washers 410 and 418, causing quick-release member 402 (e.g., UQD) of connector 400 to move radially. For any axial misalignment, the misalignment is corrected by axial movement of quick-release member 402 (e.g., UQD) of connector 400 via axial movement of piston 404 relative to piston bore member 420 via coil spring 414. Therefore, connector 400 provides two degrees of freedom for quick-release member 402 (e.g., UQD) of connector 400.

[0088] Figure 13 A computer system 1300 according to at least one embodiment is illustrated. In at least one embodiment, the computer system 1300 is configured to implement the various processes and methods described in this disclosure.

[0089] In at least one embodiment, the computer system 1300 includes, but is not limited to, at least one central processing unit (“CPU”) 1302 connected to a communication bus 1310 implemented using any suitable protocol, such as PCI (“Peripheral Component Interconnect”), Peripheral Component Interconnect High Speed ​​(“PCI-Express”), AGP (“Accelerated Graphics Port”), HyperTransport, or any other bus or point-to-point communication protocol. In at least one embodiment, the computer system 1300 includes, but is not limited to, main memory 1304 and control logic (e.g., implemented in hardware, software, or a combination thereof), with data stored in main memory 1304, which may be in the form of random access memory (“RAM”). In at least one embodiment, a network interface subsystem (“network interface”) 1322 provides an interface with other computing devices and networks for receiving data from and sending data from other systems by the computer system 1300 to other systems.

[0090] In at least one embodiment, the computer system 1300 includes, but is not limited to, an input device 1308, a parallel processing system 1312, and a display device 1306, which may be implemented using conventional cathode ray tube (“CRT”), liquid crystal display (“LCD”), light-emitting diode (“LED”), plasma display, or other suitable display technologies. In at least one embodiment, user input is received from the input device 1308, such as a keyboard, mouse, touchpad, microphone, etc. In at least one embodiment, each of the above modules may reside on a single semiconductor platform to form the processing system.

[0091] In at least one embodiment, a computer program in the form of machine-readable executable code or computer control logic algorithms is stored in main memory 1304 and / or secondary storage devices. If the computer program is executed by one or more processors, the computer program will enable system 1300 to perform various functions according to at least one embodiment. Memory 1304, storage devices, and / or any other storage devices are possible examples of computer-readable media. In at least one embodiment, secondary storage devices can refer to any suitable storage device or system, such as hard disk drives and / or removable storage drives, such as floppy disk drives, magnetic tape drives, optical disk drives, digital versatile optical disc (“DVD”) drives, recording devices, Universal Serial Bus (“USB”) flash memory, etc. In at least one embodiment, the architecture and / or functionality shown in the various figures above are implemented in the context of: CPU 1302; parallel processing system 1312; integrated circuits capable of implementing at least a portion of the functions of both CPU 1302 and parallel processing system 1312; chipsets (e.g., a set of integrated circuits designed to operate as a unit for performing various related functions and sold, etc.); and any suitable combination of integrated circuits.

[0092] In at least one embodiment, the architecture and / or functionality of the foregoing figures are implemented within the context of a general-purpose computer system, a circuit board system, a game console system dedicated to entertainment, a special-purpose system, etc. In at least one embodiment, the computer system 1300 may take the form of a desktop computer, laptop computer, tablet computer, server, supercomputer, smartphone (e.g., wireless handheld device), personal digital assistant (“PDA”), digital camera, vehicle, head-mounted display, handheld electronic device, mobile phone device, television, workstation, game console, embedded system, and / or any other type of logic.

[0093] In at least one embodiment, the parallel processing system 1312 includes, but is not limited to, multiple parallel processing units (“PPUs”) 1314 and associated memory 1316. In at least one embodiment, the PPUs 1314 are connected to a host processor or other peripheral devices via interconnects 1318 and switches 1320 or multiplexers. In at least one embodiment, the parallel processing system 1312 distributes computational tasks to parallelizable PPUs 1314, for example, as part of a thread block that distributes computational tasks to multiple graphics processing units (“GPUs”). In at least one embodiment, some or all of the PPUs 1314 share memory and are accessible (e.g., for read and / or write access), but this shared memory may result in performance penalties compared to using local memory and registers residing on the PPUs 1314. In at least one embodiment, the operation of the PPUs 1314 is synchronized using commands such as _syncthreads(), where all threads in a block (e.g., threads executing on multiple PPUs 1314) must reach a specific point of execution in the code before continuing execution.

[0094] Data centers and other network systems can include connections between data centers, switching systems, servers, racks, and devices to provide signal transmission between one or more of these components. These connections can be implemented using assemblies of cables, transceivers, interconnects, interposers, and connectors. For high-bandwidth applications and / or long-distance connections, high-power optical communication may be preferred to ensure the integrity of signal transmission.

[0095] Data centers may include multiple network switches with specific topologies, such as fat tree topology, thin-fly topology, dragonfly topology, multi-level fat tree, fat fly, Benes, Clos, mesh networks, and ring networks. The specifications and composition of the network switches in the topology will affect the overall network performance of the data center (e.g., bandwidth capacity).

[0096] Data centers are the central hubs for internet storage and data processing. The massive deployment of cloud applications has led to an exponential increase in data center size, spurring the development of faster switches to handle the ever-increasing data traffic within data centers. Currently, the most advanced switches utilize application-specific integrated circuits (ASICs) in their form, equipped with 256 data channels, each operating at 50Gb / s, enabling them to handle 12.8Tb / s of traffic. Such switching ASICs typically consume up to 400 watts of power, comparable to the power consumption of the optical transceiver interfaces attached to each ASIC. To meet the ever-increasing traffic demands, switch capacity roughly doubles every two years. To date, this rapid expansion has been made possible by leveraging advancements in manufacturing technologies, such as CMOS technology, which are collectively described by Moore's Law (i.e., the number of transistors in densely packed integrated circuits is observed to roughly double every two years). However, in recent years, there have been strong indications that Moore's Law is slowing down, raising concerns about the ability to sustain the target expansion speed of switch capacity. Therefore, alternative technologies are being researched.

[0097] As mentioned above, data centers, high-performance computing clusters, etc., are typically composed of various computing components or networked devices, while communication networks, consisting of electrical and / or optical equipment, can be used to enable communication between the networked devices that constitute these implementations. See also Figures 14A-14B For example, network architecture 1400 may include data center 1402, communication network 1404, and network device 1406. Network architecture 1400 may illustrate a general computing architecture in which more specific systems and / or subsystems may run. Although the following description refers to network architecture 1400 and / or data center 1402 that can implement embodiments of this disclosure, this disclosure contemplates that the transceiver resilient devices and techniques described herein can be applied without limitation to any communication implementation.

[0098] For example, data center 1402 may be a centralized facility designed to house computing resources and related components. Data center 1402 may operate to support the infrastructure required for advanced computing tasks, enabling efficient, secure, and reliable operation. Data center 1402 may include building and structural components, including power, cooling systems, fire protection systems, and physical security measures configured to maintain optimal operating conditions and / or protect equipment from environmental hazards and unauthorized access. Example data center 1402 may include high-performance servers or compute nodes, typically arranged in racks (e.g.,...). Figure 14BThese servers, as shown, are connected via high-speed networks as described herein. They may include processors (e.g., central processing units (CPUs), graphics processing units (GPUs), data processing units (DPUs), etc.), memory (e.g., RAM), and storage solutions (e.g., hard disk drives (HDDs), solid-state drives (SSDs), etc.). The hardware configuration may be designed for parallel processing and high throughput to meet the demands of high-performance computing (HPC) applications.

[0099] Data center 1402 may include high-speed network devices, such as network switches, routers, firewalls, etc., to facilitate fast and secure data transmission within data center 1402 (e.g., between servers or compute nodes) and with external networks. Data center 1402 can facilitate communication between servers or compute nodes through its network topology, thereby ensuring efficient data exchange, minimizing latency, and maximizing bandwidth. The network topology determines how various network devices (e.g., switches and routers) interconnect to enable data flow. By implementing an effective network topology, data center 1402 can support high-performance computing tasks. Examples of various network topologies may include hierarchical networking topologies, such as fat tree topologies, Slim Fly topologies, Dragonfly topologies, etc.

[0100] Communication network 1404 can communicatively couple data center 1402 with network device 1406 and other external devices to enable data exchange and connectivity. Examples of communication network 1404 may include Internet Protocol (IP) networks, Ethernet networks, InfiniBand (IB) networks, Fibre Channel networks, the Internet, cellular communication networks, wireless communication networks, and combinations thereof (e.g., Fibre Channel over Ethernet), variations thereof, and / or similar networks. Communication network 1404 can incorporate a variety of network types and configurations, enabling data center 1402 to accommodate a wide range of application requirements, from general data communications to dedicated HPC tasks. As described herein, communication network 1404 can utilize various optical components to establish communication links (e.g., communicatively couple) between components in architecture 1400. Therefore, communication network 1404 may include various optical devices, transceivers, modules, etc., configured to generate optical signals (e.g., provide optical transmitter functionality) and / or receive optical signals (e.g., provide optical receiver functionality).

[0101] Network device 1406 may include a variety of computing devices capable of sending and receiving signals via communication network 1404. The range of network device 1406 can range from personal computing devices to complex server configurations. Examples include personal computers (PCs), laptops, tablets, smartphones, and servers. Network device 1406 facilitates user interaction with data center 1402, allowing data to be input, retrieved, and processed from remote locations. In addition to individual computing devices, network device 1406 may also include server clusters or other data centers. For example, these could be other data centers similar to or identical to data center 1402. This interconnection allows for the construction of distributed computing environments, thereby improving redundancy, load balancing, and disaster recovery capabilities. By linking multiple data centers, network architecture 1400 can leverage geographically dispersed resources, optimize performance, and ensure high availability.

[0102] As described herein, data center 1402 and / or network device 1406 may include storage devices and processing circuitry for performing computing tasks, such as controlling internal data flow and data flow through communication network 1404. The processing circuitry may include software, hardware, or a combination thereof. For example, the processing circuitry may include memory containing executable instructions and a processor (e.g., a microprocessor) that executes those instructions. Memory may correspond to any suitable type of memory device or set of memory devices configured to store instructions. Non-limiting examples of suitable memory devices include flash memory, random access memory (RAM), read-only memory (ROM), variations thereof, combinations thereof, or similar technologies. In specific embodiments, the memory and processor may be integrated into a general-purpose device, such as a microprocessor with integrated memory. Additionally or alternatively, the processing circuitry may include hardware components such as application-specific integrated circuits (ASICs). Other non-limiting examples of processing circuitry include integrated circuit (IC) chips, CPUs, GPUs, microprocessors, field-programmable gate arrays (FPGAs), sets of logic gates or transistors, resistors, capacitors, inductors, and diodes. Part or all of the processing circuitry can be integrated on a printed circuit board (PCB) or an assembly of PCBs. It should be understood that any suitable type of electrical component or assembly of electrical components may be suitable for inclusion in the processing circuitry.

[0103] Furthermore, although not explicitly shown, this disclosure envisions that data center 1402 and network device 1406 may include one or more communication interfaces to facilitate wired and / or wireless communication between themselves and with other components not shown in network architecture 1400. These communication interfaces may include various technologies, including but not limited to Ethernet ports, fiber optic connections, Wi-Fi® transceivers, Bluetooth® modules, and cellular communication modules, to enable integration and interoperability between the various components in network architecture 1400.

[0104] Furthermore, this disclosure envisions that network architecture 1400 may include other components and functions. For example, the network architecture may include, but is not limited to, additional processing units, dedicated accelerators (e.g., tensor processing units or TPUs), enhanced security modules, and redundant power supplies. Including these elements can be designed to ensure that network architecture 1400 is robust, scalable, and capable of meeting various operational requirements. Any variations, modifications, or adjustments to the described elements that fall within the spirit and scope of this disclosure should be considered as included within this disclosure. This includes any combination, sub-combination, or enhancement of the various described elements to achieve improved performance, reliability, and efficiency in network architecture 1400.

[0105] In high-capacity data center networks (e.g.) Figures 14A-14B In the network shown, communication network 1404 can utilize optical transceivers that send and receive optical signals via optical fiber or other optical communication media to establish connections between devices in architecture 1400.

[0106] In at least one exemplary embodiment, data center 1402 corresponds to a set of network devices, such as network switches (e.g., Ethernet switches, IP routers, multi-service platforms, various transport network elements, traditional communication equipment, or any other suitable communication system), which are connected to a collection of servers or computing nodes.

[0107] A switching fabric is used to transmit data between switch ports. A switching fabric includes one or more interconnecting circuits, which can be arranged using various switching fabric architectures, such as... Crossbar switches, Banyan, Benes, Omega, Clos, multiplane, STS, TST, shared memory, buffered crossbar switches, any other suitable blocking or non-blocking architecture, or any suitable hybrid architecture thereof. In a typical embodiment, the switching structure is implemented in hardware, which may include field-programmable gate arrays (FPGAs) and / or application-specific integrated circuits (ASICs), and in some implementations may also include bus interconnects. Data center 1402 may employ a network topology (e.g., a hierarchical network topology), such as a fat tree topology, a slim fly topology, a dragonfly topology, etc. Data center 1402 routes traffic between network switches and servers within its premises, and at least one layer of the topology of data center 1402 is coupled to communication network 1404 to allow network traffic to flow between data center 1402 and network devices 1406.

[0108] Communication network 1404 can communicatively couple data center 1402 with network device 1406 and other external devices to enable data exchange and connectivity. Examples of communication network 1404 may include Internet Protocol (IP) networks, Ethernet networks, InfiniBand (IB) networks, Fibre Channel networks, the Internet, cellular communication networks, wireless communication networks, combinations thereof (e.g., Fibre Channel over Ethernet), variations thereof, and / or similar networks.

[0109] Figure 15The illustration depicts an example network configuration 1500, whose components can be used to implement various aspects of different embodiments, such as providing, generating, modifying, encoding, processing, fusing, and / or transmitting generated image data, calculated measurements, or other such content. In at least one embodiment, client device 1502 can use components of content application 1504 on client device 1502, along with data locally stored on the client device, to generate or receive data for a session. In at least one embodiment, content application 1524 executing on a computer or processor 1520 (e.g., a cloud server or control system) can initiate a session associated with at least one client device 1502 (e.g., a vehicle or robot), such as using a session manager and user data stored in user database 1536, and can enable the selection and / or retrieval of content such as liquid coolant or server thermal data from storage 1534 for use by test module 1532 to calculate one or more performance metrics for monitoring module 1528, which can provide flow or thermal data to control the flow or temperature in an environment where the data is to be used to determine the appropriate operating conditions. Content manager 1526 can work with these different modules to perform tests and analyses and may instruct any actions to be taken in response to performance metrics failing to meet operational requirements. At least a portion of the data or instruction content can be transmitted to client device 1502 and / or physical device 1570 using a suitable transmission manager 1522 for sending via download, streaming, or other such transmission channels. An encoder can be used to encode and / or compress at least a portion of the data before it is transmitted to client device 1502. In at least one embodiment, client device 1502 receiving such content can provide it to a corresponding content application 1504, which may also or optionally include a graphical user interface 1510, a traffic monitoring module 1512, and a control module 1514 for providing, compositing, rendering, combining, modifying, or using the content on or through client device 1502 for presentation, navigation, control (or other purposes), such as transmission to physical device 1570. In some embodiments, the computer / processor 1520 and the client device 1502 may be able to communicate directly without transmitting data over the network 1540 to avoid issues such as latency and availability. A decoder may also be used to decode data received over the network 1540 for presentation by the client device 1502, such as presenting image content or performance metrics via a display device 1506, and playing audio (e.g., corresponding sounds or synthesized speech) via at least one audio playback device 1508 (e.g., a speaker or headphones).In at least one embodiment, at least a portion of the content may already be stored on, rendered on, or accessible to client device 1502, so that at least a portion of the content does not need to be transmitted over network 1540. For example, the content (e.g., hot data) may have been pre-downloaded or stored locally on a hard disk or optical disc. In at least one embodiment, the content may be transmitted from computer / processor 1520 or user database 1536 to client device 1502 using a transmission mechanism such as data streaming. In at least one embodiment, at least a portion of the content may be obtained, enhanced, and / or streamed from other sources (e.g., third-party services 1560 or other client devices 1550), which may also include content applications for generating, updating, enhancing, or providing map content. In at least one embodiment, a portion of this functionality may be performed using multiple computing devices, or multiple processors within one or more computing devices, for example, a combination of CPU and GPU (graphics processing unit).

[0110] In at least some of these examples, the client device may include any suitable computing device, such as a desktop computer, laptop, set-top box, streaming media device, game console, smartphone, tablet, VR headset, AR glasses, wearable computer, or smart TV. Each client device may submit requests via at least one wired or wireless network, such as the Internet, Ethernet, local area network (LAN), or cellular network. In this example, these requests may be submitted to an address associated with a cloud provider that operates or controls one or more electronic resources within the cloud provider environment, such as a data center or server farm. In at least one embodiment, the request may be received or processed by at least one edge server located at the network edge and outside at least one security layer associated with the cloud provider environment. This reduces latency and improves the security of resources within the cloud provider environment by allowing client devices to interact with a closer server.

[0111] In at least one embodiment, the system can be used to monitor or manage the thermal status of a server that includes a cold plate as a liquid manifold. In other embodiments, the system can be used for other purposes, such as providing control over the flow of liquid coolant or performing deep learning operations. In at least one embodiment, the system can be implemented using an edge device or may include one or more virtual machines (VMs). In at least one embodiment, the system can be implemented at least partially in a data center or at least partially using cloud computing resources.

[0112] In at least one embodiment, the cold plate includes adjustable fins that form microchannels through which a heat transfer fluid flows. In at least one embodiment, the fins in the cold plate are capable of transferring heat from at least one associated computing device to the heat transfer fluid flowing through the microchannels formed between the multiple fins. In at least one embodiment, the fins of the cold plate can be dynamically adjusted in real time to allow more heat from at least one computing device to be transferred to the heat transfer fluid flowing through the finned cold plate. In at least one embodiment, such fins can be adjusted by a processor or processorless system in part based on a temperature determined (e.g., sensed) for the cold plate. In at least one embodiment, the temperature can be associated with at least one computing device, the workload of at least one computing device, or the heat transfer fluid at different time periods and at the inlet and outlet of the cold plate. In at least one embodiment, the processorless system can rely on the thermal properties of at least two materials used to form the cold plate fins such that these fins can react without a processor, thereby exposing more surface area to the heat transfer fluid. In at least one embodiment, these fins can include overlapping portions, which can be exposed by the action of a control mechanism or by the properties of the at least two materials that together constitute the fins.

[0113] In at least one embodiment, the following can be used: Figure 16 The exemplary data center 1600 shown herein has a cooling system that incorporates the improvements described herein. In at least one embodiment, numerous specific details are set forth herein for detailed understanding, but the concepts presented herein can be practiced even without considering one or more of these specific details. In at least one embodiment, the data center cooling system can respond to sudden bursts of high heat demands caused by the constantly changing computing loads in today's computing components. In at least one embodiment, because these demands vary, or the range of cooling demands may vary from minimum to maximum, a suitable cooling system must be used to meet these demands economically. In at least one embodiment, a liquid cooling system can be used for medium to high cooling demands. In at least one embodiment, partial immersion cooling can economically meet higher cooling demands. In at least one embodiment, these different cooling demands also reflect different thermal characteristics of the data center. In at least one embodiment, the heat generated by these components, servers, and racks is collectively referred to as thermal characteristics or cooling demands, because cooling demands must fully address thermal characteristics.

[0114] In at least one embodiment, a data center liquid cooling system is disclosed. In at least one embodiment, this data center cooling system can address the thermal characteristics of associated computing or data center equipment, such as graphics processing units (GPUs), switches, dual in-line memory modules (DIMMs), or central processing units (CPUs). In at least one embodiment, these components may be referred to herein as high heat-density computing components. Furthermore, in at least one embodiment, the associated computing or data center equipment may be a processing card having one or more GPUs, switches, or CPUs thereon. In at least one embodiment, each of the GPU, switch, and CPU can serve as a thermal characteristic of the computing device. In at least one embodiment, the GPU, CPU, or switch may have one or more cores, and each core can serve as a thermal characteristic.

[0115] In at least one embodiment, the cold plate includes adjustable fins that form microchannels through which a heat transfer fluid flows. In at least one embodiment, the fins in the cold plate are capable of transferring heat from at least one associated computing device to the heat transfer fluid flowing through the microchannels formed between the multiple fins. In at least one embodiment, the fins of the cold plate can be dynamically adjusted in real time to transfer more heat from at least one computing device to the heat transfer fluid flowing through the finned cold plate. In at least one embodiment, such fins may be adjusted by a processor or processorless system in part based on a temperature determined (e.g., sensed) for the cold plate. In at least one embodiment, the temperature may be associated with at least one computing device, the workload of at least one computing device, or the heat transfer fluid at different time periods and at the inlet and outlet of the cold plate. In at least one embodiment, the processorless system may rely on the thermal properties of at least two materials used to form the fins of the cold plate such that the fins can react without a processor, thereby exposing more surface area to the heat transfer fluid. In at least one embodiment, the fins may include overlapping portions, which may be exposed by the action of a control mechanism or by the properties of the at least two materials that together constitute the fins.

[0116] In at least one embodiment, the cold plate has a top plate, a bottom plate, and fins located between the two. In at least one embodiment, the bottom plate may be a base for the cold plate. In at least one embodiment, the top plate may be an intermediate portion between a cover plate of the cold plate and a bottom plate or substrate of the cold plate. In at least one embodiment, the fins may be coupled to the bottom plate or substrate and the top plate such that the top plate is movable to expose the overlapping portion of each fin and expose the overlapping portion of each fin to heat transfer fluid flowing through the cold plate. In at least one embodiment, the exposure of the overlapping portion of each fin exposes a previously covered surface area to the heat transfer fluid, thereby providing additional cooling to the fins of the cold plate and, by association, additional cooling to the associated computing device.

[0117] In at least one embodiment, multiple fins form microchannels for the flow of heat transfer fluid therethrough. In at least one embodiment, these fins are capable of actively or passively responding to thermal feedback by altering the microchannels, thereby enabling the heat transfer fluid to absorb more heat from at least one computing device. In at least one embodiment, the active response can be implemented by at least one processor that can expose more surface area of ​​such fins by expanding the overlapping portions of such fins. In at least one embodiment, the passive response can be implemented by the thermal properties of the material constituting each fin in the fins together, which allow such fins to expand.

[0118] In at least one embodiment, this paper addresses the problem of cold plates being static devices through a smart dynamic cold plate. In at least one embodiment, the smart dynamic cold plate (through its internal structure) enables the cold plate to respond to temperatures sensed or determined from at least one computing device. In at least one embodiment, the intelligent aspects of the smart dynamic cold plate, compared to a static cold plate, allow for the adjustment of the fins of such a cold plate using sensor input. In at least one embodiment, the microchannels formed by these fins can be modified, thereby altering the fluid path or increasing the interaction area between the heat transfer fluid and each fin. In at least one embodiment, these aspects allow more heat transfer fluid to flow through a specific area, or to flow through a specific area and allow for heat removal in those areas where the at least one computing device generates high heat density.

[0119] In at least one embodiment, a static cold plate for liquid cooling of GPUs, CPUs, switches, and other high heat density components may have static microchannels, allowing heat transfer fluids to flow therethrough to remove heat from such heat dissipation components in a data center. In at least one embodiment, the static cold plate employs a heat removal design and method that is independent of or does not respond to the heat density or heat dissipation of the computing components. In at least one embodiment, the thermal behavior of some cold plates may vary due to computing, environmental, and other properties that require dynamic behavior to achieve optimal heat removal capabilities for available resources in a liquid-cooled environment.

[0120] In at least one embodiment, the liquid-cooled indirect cooling plate may consist of components therein. In at least one embodiment, two components may be provided to provide a rigid mechanical attachment for the bottom or base layer or component and to function as a highly thermally conductive medium, wherein heat is conducted from computing components (GPUs, switches, and CPUs) to multiple fins forming microchannels on the bottom or base plate or component. In at least one embodiment, these fins are built into a base metal and may be associated with a top or upper plate or component that serves as an intermediate plate or component. In at least one embodiment, the top or upper plate or component that serves as the intermediate plate or component is non-conductive and is made of a non-conductive material. In at least one embodiment, a cover plate (along with side plates) encloses these components within the smart dynamic cooling plate.

[0121] In at least one embodiment, the top or upper plate or component, through the function of its intermediate plate or component, can dynamically alter the microchannel fluid path of the substrate, partially based on the instantaneous thermal behavior of the heating components associated with the smart dynamic cold plate. In at least one embodiment, utilizing intelligent sensing, inference, and adaptive modification, the microchannels forming the fluid path can be dynamically adjusted to direct more heat transfer fluid (for heat removal) to or flow towards areas of high heat density within the smart dynamic cold plate. In at least one embodiment, these features also allow for simultaneous reduction of fluid flow, thereby blocking microchannels in areas with lower heat removal requirements, which are determined or sensed from the substrate of the smart dynamic cold plate. In at least one embodiment, the overlapping portions of the fins provided within the smart dynamic cold plate can block or restrict microchannels by thickening the overlapping portions between the fins of the smart dynamic cold plate. In at least one embodiment, multiple top plates can be provided as intermediate plates, and each top plate can be associated with different fins. In at least one embodiment, movement of different top plates can achieve different blocking, restriction, or airflow redirection within the smart dynamic cold plate.

[0122] In at least one embodiment, the following can be used: Figure 16The exemplary data center 1600 shown herein has a cooling system that can be improved as described herein. In at least one embodiment, the data center 1600 may be one or more server rooms 1602, each housing racks 1610 and auxiliary equipment for accommodating one or more servers on one or more server bays. In at least one embodiment, the data center 1600 is supported by a cooling tower 1604 located outside the data center 1600. In at least one embodiment, the cooling tower 1604 dissipates heat from inside the data center 1600 by acting on a main cooling loop 1606. In at least one embodiment, a cooling distribution unit (CDU) 1612 is provided between the main cooling loop 1606 and a second or auxiliary cooling loop 1608 to absorb heat from the second or auxiliary cooling loop 1608 back to the main cooling loop 1606. In at least one embodiment, in one aspect, the auxiliary cooling loop 1608 may have various pipes connected to the server bays as needed. In at least one embodiment, loops 1606 and 1608 are shown in line diagram form, but those skilled in the art will understand that one or more pipe structures may be used. In at least one embodiment, flexible polyvinyl chloride (PVC) pipes and associated piping can be used to move the heat transfer fluid in each provided loop 1606, 1608. In at least one embodiment, one or more coolant pumps can be used to maintain the pressure difference within the coolant loops 1606, 1608 to move the coolant based on temperature sensors located at different locations, including within a server room, within one or more racks 1610, and / or within server chassis or server racks within one or more racks 1610.

[0123] In at least one embodiment, the coolant in the main cooling circuit 1606 and the auxiliary cooling circuit 1608 may be at least water and an additive. In at least one embodiment, the additive may be ethylene glycol or propylene glycol. In operation, in at least one embodiment, the main cooling circuit and the auxiliary cooling circuit may each have their own coolant. In at least one embodiment, the coolant in the auxiliary cooling circuit may be dedicated to the needs of components in the server bay or associated rack 1610. In at least one embodiment, the CDU 1612 is capable of fine control over the coolant provided in the coolant circuits 1606, 1608, either independently or simultaneously. In at least one embodiment, the CDU may be adapted to control the flow rate of the coolant so that the coolant is properly distributed to absorb heat generated within the associated rack 1610. In at least one embodiment, a more flexible conduit 1614 is provided from the auxiliary cooling circuit 1608 to each server bay to provide coolant to the electrical and / or computing components therein.

[0124] Heat transfer fluids typically include water, aqueous solutions (e.g., propylene glycol-water), brine, antifreeze, mixtures of antifreeze and water, oil, alcohol, mercury or the like, or any other suitable heat-conducting fluid. The heat transfer fluid can be a conductive coolant and can include water, deionized water, or a refrigerant (e.g., R-134a), a mixture of water and additives (e.g., a mixture of water and ethylene glycol or a mixture of water and propylene glycol, such as 25% propylene glycol in deionized water). The heat transfer fluid can also be a single dielectric fluid (e.g., water-free for the purposes of this disclosure), or water combined with an additive containing at least one dielectric fluid, such as one or more of deionized water, ethylene glycol, and propylene glycol. In at least one embodiment, the heat transfer fluid can be an absorption chiller whose working fluid is a mixed solution containing lithium bromide as the absorbent material and water as the carrier material. The heat transfer fluid can also be a two-phase coolant with a boiling point below the expected operating temperature of the electronic device. Exemplary two-phase coolants include 2,3,3,3-tetrafluoropropylene, 1,1,1,2-tetrafluoroethane, and water.

[0125] In at least one embodiment, the conduit 1618 forming part of the auxiliary cooling circuit 1608 may be referred to as a server room manifold. Furthermore, in at least one embodiment, an additional conduit 1616 may extend from the manifold conduit 1618 and may also form part of the auxiliary cooling circuit 1608, but may be referred to as a manifold. In at least one embodiment, the coolant conduit 1614 enters the rack as part of the auxiliary cooling circuit 1608, but may be referred to as a rack cooling manifold within one or more racks. In at least one embodiment, the manifold 1616 extends to all racks along a row in the data center 1600. In at least one embodiment, the piping of the auxiliary cooling circuit 1608, including coolant manifolds 1618, 1616, and 1614, may be improved by at least one embodiment herein. In at least one embodiment, a cooler 1620 may be provided in the main cooling circuit within the data center 1602 to support cooling before the cooling tower. In at least one embodiment, an additional cooling circuit, which may exist in the main control circuit and provide external cooling to both the rack exterior and the auxiliary cooling circuit, may be integrated with the main cooling circuit and distinguished from the auxiliary cooling circuit in this disclosure.

[0126] In at least one embodiment, during operation, heat generated within the server trays of the provided rack 1610 can be transferred via the flexible conduit of the manifold 1614 of the second cooling circuit 1608 to the coolant leaving one or more racks 1610. In at least one embodiment, a second coolant from CDU 1612 (in the auxiliary cooling circuit 1608) is used to cool the provided rack 1610 and flows through the provided conduit to one or more racks 1610. In at least one embodiment, the second coolant from CDU 1612 flows out from one side of a data center manifold having conduit 1618, flows through manifold 1616 to one side of the rack 1610, and flows through another conduit 1614 through one side of the server tray. In at least one embodiment, used or returned second coolant (or departing second coolant carrying heat from the computing components) flows out from the other side of the server tray (e.g., entering from the left side of the rack and exiting from the right side of the server tray after circulating through the server tray or the components on the server tray). In at least one embodiment, the used second coolant exiting the server tray or rack 1610 flows out from a different side (e.g., the exit side) of the conduit 1614 and enters a parallel but also exit side of the manifold 1616. In at least one embodiment, the used second coolant enters from the manifold 1616 into a parallel portion of the data center manifold 1618, flowing in the opposite direction to the incoming second coolant (which may also be a renewed second coolant), and flows toward the CDU 1612.

[0127] In at least one embodiment, the used second coolant exchanges heat with the main coolant in the main cooling circuit 1606 via CDU 1612. In at least one embodiment, the used second coolant can be refreshed (e.g., at a relatively lower temperature compared to the temperature of the used second coolant stage) and prepared to be circulated back to one or more computing components via the second cooling circuit 1608. In at least one embodiment, various flow and temperature control features in CDU 1612 are capable of controlling the heat exchanged from the used second coolant, or controlling the flow rate of the second coolant into and out of CDU 1612. In at least one embodiment, CDU 1612 is also capable of controlling the flow rate of the main coolant in the main cooling circuit 1606.

[0128] In one embodiment, the switch may be a 1U switch, where "1U" refers to the industry-standard size of rack-mounted switches and servers. The switch may be an electrical switch, an optical switch, a hybrid electro-optical switch, or any combination thereof. The switch may be implemented using suitable hardware and / or software to enable signal routing within the appropriate domain. For example, an electrical switch may include a receiver that receives optical signals and converts them into electrical signals for routing within the electrical switch. The receiver of an electrical switch may include a transimpedance amplifier (TIA), a photodetector, and a controller, which together are used to convert optical signals into electrical signals. Each electrical switch may also include a transmitter that converts electrical signals routed internally within the electrical switch into optical signals for output to other switches (optical switches or electrical switches) within the system. For example, the transmitter of an electrical switch may include a light source, a modulator, and a controller, with the controller controlling the modulator and the light source. In some embodiments, the receiver / transmitter pair may be integrated into a single transceiver. Each electrical switch may also include an internal switching circuitry system for routing electrical signals within the electrical switch.

[0129] Throughout this specification, the terms "electrical switch," "electrical switching ASIC," "ASIC," and variations thereof are used interchangeably. While an electrical switch can be implemented using an ASIC, the embodiments are not limited thereto, and an electrical switch can be implemented using any suitable hardware and / or software capable of routing signals in the electrical domain.

[0130] For example, an electrical switch may include a receiver that receives optical signals and converts them into electrical signals for routing within the switch. For example, the receiver of an electrical switch may include a transimpedance amplifier (TIA), a photodetector, and a controller, which together are used to convert the optical signals into electrical signals. Each electrical switch may also include a transmitter that converts the electrical signals routed within the electrical switch into optical signals for output to other switches (optical switches or electrical switches) in the system. For example, the transmitter of an electrical switch may include a light source, a modulator, and a controller that controls the modulator and the light source. In at least one exemplary embodiment, the receiver / transmitter pair is integrated into a single transceiver. Each electrical switch may also include internal switching circuitry for routing electrical signals within the electrical switch.

[0131] High-capacity optical switch assemblies exchange data across multiple channels at high data rates, with up to hundreds of channels and data rates reaching hundreds of Gb / s (Gb / s = 10⁹ bits / second). To save power, it is preferable to co-package the switch itself with the "optical engine." The optical engine is typically a small, high-density optical transceiver located within an application-specific integrated circuit (ASIC), or integrated with the switch within an ASIC package.

[0132] The switch assembly is housed in a rack-mounted chassis with optical sockets on its front panel for easy access. Fiber optic cables are used to transmit signals to and from the ASIC to the optical sockets.

[0133] To further simplify installation and use, it is sometimes necessary to detach the fiber optic cable from the transceiver so that a thinner cable can be routed using a mounting device. Each fiber optic cable can be designed to work with a specific transceiver, rather than including the transceiver itself. The transceiver connects to a node (e.g., a server) and is used to connect the connectors for each cable to that node, as described herein.

[0134] In one embodiment, the optical switch assembly may include one or more processing circuits; the processing circuits may include firmware (FW).

[0135] Optical switches are a solution driving network technology advancement due to their high data capacity and low power consumption potential. An optical switch has optical input and output ports, capable of routing light coupled to the input port to the target output port on demand, based on one or more control signals (electrical or optical control signals). Signal routing is performed in the optical domain, eliminating the need for photoelectric and electro-optical conversions, thus avoiding the use of power-intensive transceivers. Because header processing and data buffering cannot be performed in the optical domain, packet switching (as implemented in electrical switches) is not possible. Instead, a circuit-switching paradigm is used: communication is achieved by establishing an end-to-end circuit between the two endpoints connected to the input and output of the optical switch. Optical switches can be used in the most common data center interconnect topologies, such as Fat Tree, SlimFly, and Dragonfly+.

[0136] Optical switches may include hardware and / or software for routing signals in the optical domain. Thus, in one embodiment, an optical switch may include input and output optical fibers carrying optical signals, and one or more devices suitable for routing the optical signals within the optical switch. For example, one or more devices for routing the optical signals may include one or more movable mirrors (e.g., MEMS mirrors) controlled to move in a manner that guides light from the input fiber to the desired output fiber, or to force or guide light from one waveguide to another. Optical switches may include one or more devices for amplifying light to compensate for propagation and scattering losses introduced by the optical switch. In at least one example embodiment, both the input and output signals of the ASIC are optical signals, meaning that each optical switch connected to an electrical switch routes the optical signal received from the electrical switch without the need for hardware and / or software to convert electrical signals to optical signals for routing within the optical switch. However, the example embodiments are not limited to this; if required (e.g., if the input and / or output signals are electrical signals), the optical switch may include electro-optical-electrical conversion hardware and / or software.

[0137] In one embodiment, the optical switch may include an arrayed waveguide grating router (AWGR), which is a passive switch structure. In some embodiments, the optical switch may correspond to a passive element that operates as a wavelength router, which interconnects the output and input using multiple wavelengths by following a specific cyclic wavelength routing pattern.

[0138] On the other hand, optical switches can operate by directly routing optical signals without converting them to electrical signals. Each optical switch may include optical receivers, such as photodetectors and wavelength division multiplexing (WDM) demultiplexers, which receive incoming optical signals. These optical signals are then guided by internal optical switching components, such as microelectromechanical systems (MEMS) mirrors, waveguides, or optical cross-connects, which route the signals to the appropriate output path. Optical switches may also include optical transmitters, such as laser diodes and modulators, which transmit the routed optical signals to the next switch in the network. Hybrid electro-optical switches can combine electrical and optical components to route signals. Such switches may include receivers that use TIAs and photodetectors to convert optical signals into electrical signals, similar to receivers in electrical switches. These electrical signals can then be routed within the switch using an internal electrical switching circuitry. Furthermore, hybrid switches may include optical switching components, such as WDM multiplexers and MEMS devices, to directly route optical signals. The transmitters in a hybrid switch may include both electro-optical converters and direct optical transmitters, enabling the hybrid switch to interface with both electrical and optical networks. For example, the transmitter of a hybrid switch may include a light source, an optical signal modulator, and a traditional electrical signal transmitter, thereby providing routing capabilities across different signal domains.

[0139] Interconnections between switches in a network topology can be achieved via fiber optic cables or traditional cables, depending on the specific requirements of the system. For example, communication channels can consist of dedicated differential cable pairs and / or fiber optic cables, each customized to provide optimal performance for data transmission needs. The dedicated differential cable pairs used in these interconnects can contain various cable media, such as copper, aluminum, gold, silver, nickel, or composite materials, such as copper-clad aluminum, copper-clad steel, or bimetallic conductors. The selection of these materials should consider their conductivity and durability to ensure reliable and efficient data transmission. For example, in a four-channel network, each channel can consist of its own dedicated copper cable, providing an isolated physical path for each communication channel to deserialize the data stream. This configuration helps maintain signal integrity and reduce crosstalk between channels.

[0140] Alternatively, fiber optic cables can be used for interconnection. Optical fibers can transmit data streams through different wavelengths of light, each assigned a unique wavelength. Using fiber optic cables allows multiple data streams to be transmitted simultaneously over a single fiber, significantly improving network bandwidth and efficiency, especially suitable for long-distance data transmission and applications requiring high data transmission rates. Various optical networking technologies can be used to transmit multiple optical signals (e.g., data signals or data streams) over a single optical fiber in an optical link with minimal optical signal interference. These technologies can improve bandwidth efficiency and reduce the amount of infrastructure required for data communication.

[0141] One such technology is Time Division Multiplexing (TDM). In TDM, multiple optical signals can be transmitted over a single optical fiber by assigning a corresponding time slot to each optical signal and transmitting the signal within the assigned time slot. The time slots are allocated cyclically, with each optical signal transmitting a small amount of data within its assigned time slot. The time slots are very short, on the order of microseconds, and the cycle repeats multiple times per second, thus enabling fast data transmission.

[0142] Another technology is Frequency Division Multiplexing (FDM). In FDM, multiple optical signals can be transmitted over a single optical fiber by assigning a corresponding frequency band to each optical signal. Each optical signal is modulated onto a corresponding carrier frequency to generate a modulated signal, and these modulated signals are combined and transmitted over the single optical fiber. At the receiver, filters (such as bandpass filters) are used to separate the modulated signals. These filters allow optical signals conforming to specific frequency specifications to pass through while filtering out other signals. FDM allows an optical link to transmit multiple channels simultaneously on the same frequency band.

[0143] Another technology is wavelength division multiplexing (WDM). In WDM, multiple optical signals with different wavelengths are combined into a single optical signal and transmitted through a single optical fiber. WDM technology involves combining and separating multiple optical signals with different wavelengths onto a single optical fiber, thereby allowing more data to be transmitted and increasing the capacity of the fiber.

[0144] Examples of WDM technologies include coarse wavelength division multiplexing (CWDM) and dense wavelength division multiplexing (DWDM). CWDM combines multiple optical signals of different wavelengths into a single optical signal and transmits it through a single optical fiber. CWDM uses a wider wavelength spacing, such as about 80 nanometers (nm), which means it supports fewer channels and has a lower power budget, making it suitable for shorter distances, up to about 80 kilometers (km). CWDM requires simpler equipment and lower-cost optical components, making it a cost-effective solution for applications that do not require dense wavelength spacing. In contrast, DWDM uses a narrower wavelength spacing, such as about 0.8 nm, achieving higher channel capacity and longer distances, but is generally more expensive and structurally more complex.

[0145] In at least one embodiment, the techniques described herein are applied to active cables. Active cables are sometimes also referred to as flexible cables, smart cables, or other similar names. Active cables may include circuitry for performing functions such as signal processing, transmission format conversion, and transmission medium conversion (e.g., conversion between electrical and optical transmission media). In at least one embodiment, the active cable includes circuitry for collecting telemetry data and analyzing the data using the machine learning models described herein. Active cables may also include one or more general-purpose or special-purpose processors for using machine learning models to obtain inferences about network traffic patterns and responding to such inferences.

[0146] Figure 17 The illustration depicts a client-server network 1704 formed by a plurality of interconnected network server computers 1702 according to at least one embodiment. In at least one embodiment, each network server computer 1702 stores data that is accessible to other network server computers 1702 and client computers 1706 and networks 1708 linked to the wide area network 1704. In at least one embodiment, the configuration of the client-server network 1704 may change over time as client computers 1706 and one or more networks 1708 connect and disconnect from the network 1704, and one or more trunk line server computers 1702 are added to or removed from the network 1704. In at least one embodiment, the client-server network includes client computers 1706 and networks 1708 when they are connected to network server computers 1702. In at least one embodiment, the term "computer" includes any device or machine capable of accepting data, applying prescribed processing to the data, and providing processing results.

[0147] In at least one embodiment, the client-server network 1704 stores information accessible to the network server computer 1702, the remote network 1708, and the client computer 1706. In at least one embodiment, the network server computer 1702 is formed from a mainframe computer, a minicomputer, and / or a microcomputer, each having one or more processors. In at least one embodiment, the server computers 1702 are linked together via wired and / or wireless transmission media, such as wires, fiber optic cables, and / or microwave transmission media, satellite transmission media, or other conductive, optical, or electromagnetic wave transmission media. In at least one embodiment, the client computer 1706 accesses the network server computer 1702 via a similar wired or wireless transmission medium. In at least one embodiment, the client computer 1706 can be linked to the client-server network 1704 using a modem and a standard telephone communication network. In at least one embodiment, alternative carrier systems such as wired and satellite communication systems can also be used to link to the client-server network 1704. In at least one embodiment, other dedicated or time-sharing carrier systems can be used. In at least one embodiment, the network 1704 is a global information network, such as the Internet. In at least one embodiment, the network is a private intranet using protocols similar to the Internet but with additional security measures and restricted access controls. In at least one embodiment, network 1704 is a private or semi-private network using proprietary communication protocols.

[0148] In at least one embodiment, the client computer 1706 is any end-user computer, and may also be a mainframe computer, minicomputer, or microcomputer with one or more microprocessors. In at least one embodiment, the server computer 1702 may sometimes act as a client computer to access other server computers 1702. In at least one embodiment, the remote network 1708 may be a local area network (LAN), a network added to a wide area network by an independent service provider (ISP) via the Internet, or another group of computers interconnected via wired or wireless transmission media, the configuration of which may be fixed or change over time. In at least one embodiment, the client computer 1706 may independently or via the remote network 1708 link to and access the network 1704.

[0149] Figure 18The illustration depicts a computer network 1808 connecting one or more computers according to at least one embodiment. In at least one embodiment, network 1808 can be any type of electronically connected group of computers, including networks such as the Internet, intranets, local area networks (LANs), wide area networks (WANs), or interconnections of these network types. In at least one embodiment, connections within network 1808 can be remote modems, Ethernet (IEEE 802.3), Token Ring (IEEE 802.5), Fiber Distributed Data Link Interface (FDDI), Asynchronous Transfer Mode (ATM), or any other communication protocol. In at least one embodiment, computing devices linked to the network can be desktop computers, servers, portable devices, handheld devices, set-top boxes, personal digital assistants (PDAs), terminals, or any other desired type or configuration. In at least one embodiment, the processing power, internal memory, and other performance aspects of network-connected devices can vary considerably depending on their functionality. In at least one embodiment, communication within the network and with computing devices connected to the network can be wireless or wired. In at least one embodiment, network 1808 may at least partially comprise the global public Internet, which typically connects multiple users in a client-server model according to the Transmission Control Protocol / Internet Protocol (TCP / IP) specification. In at least one embodiment, the client-server network is the primary model for communication between two computers. In at least one embodiment, a client computer (“client”) issues one or more commands to a server computer (“server”). In at least one embodiment, the server executes client commands by accessing available network resources and returning information to the client in accordance with the client commands. In at least one embodiment, client computer systems and network resources residing on the network server are assigned network addresses for identification during communication between network elements. In at least one embodiment, communication from other network-connected systems to the server includes the network address of the relevant server / network resource as part of the communication to identify the appropriate destination of the data / request as the recipient. In at least one embodiment, when network 1808 comprises the global Internet, the network address is an IP address in TCP / IP format that can at least partially route data to email accounts, websites, or other Internet tools residing on the server. In at least one embodiment, information and services residing on a web server can be made available to a web browser on a client computer via a domain name (e.g., www.site.com) mapped to the IP address of the web server.

[0150] In at least one embodiment, multiple clients 1802, 1804, and 1806 are connected to network 1808 via their respective communication links. In at least one embodiment, each of these clients can access network 1808 via any desired communication method, such as a dial-up modem connection, cable link, digital subscriber line (DSL), wireless or satellite link, or any other communication method. In at least one embodiment, each client can communicate using any machine compatible with network 1808, such as a personal computer (PC), workstation, dedicated terminal, personal digital assistant (PDA), or other similar device. In at least one embodiment, clients 1802, 1804, and 1806 may or may not be located in the same geographical area.

[0151] In at least one embodiment, multiple servers 1810, 1812, and 1814 are connected to network 1808 to serve clients communicating with network 1808. In at least one embodiment, each server is typically a powerful computer or device that manages network resources and responds to client commands. In at least one embodiment, the server includes computer-readable data storage media, such as hard disk drives and RAM memory, which store program instructions and data. In at least one embodiment, servers 1810, 1812, and 1814 run applications that respond to client commands. In at least one embodiment, server 1810 may run a web server application to respond to client requests for HTML pages and may also run a mail server application to receive and route emails. In at least one embodiment, other applications, such as an FTP server or a media server for streaming audio / video data to clients, may also run on server 1810. In at least one embodiment, different servers may be dedicated to performing different tasks. In at least one embodiment, server 1810 may be a dedicated web server that manages resources associated with websites for various users, while server 1812 may be dedicated to providing email management. In at least one embodiment, the other servers may be dedicated to any combination of two or more services, such as media (audio, video, etc.), File Transfer Protocol (FTP), or services typically provided or available over a network. In at least one embodiment, each server may be located in the same or different location as the other servers. In at least one embodiment, multiple servers may perform mirroring tasks for users, thereby alleviating congestion or minimizing traffic flowing to and from a single server. In at least one embodiment, servers 1810, 1812, and 1814 are controlled by a web hosting provider whose business is maintaining and delivering third-party content through network 1808.

[0152] In at least one embodiment, the web hosting provider serves two different types of clients. In at least one embodiment, one type (maybe a browser) requests content, such as web pages, emails, video clips, etc., from servers 1810, 1812, 1814. In at least one embodiment, a second type (maybe a user) hires the web hosting provider to maintain network resources (e.g., a website) and make them available to the browser. In at least one embodiment, the user contracts with the web hosting provider to provide storage space, processor capacity, and communication bandwidth for the network resources they require, based on the amount of server resources they wish to use.

[0153] In at least one embodiment, in order for a web hosting provider to serve both types of clients, the application managing the web resources hosted on the server must be properly configured. In at least one embodiment, the program configuration process includes defining a set of parameters that at least partially control the application's response to browser requests and at least partially define the server resources available to a particular user.

[0154] In one embodiment, intranet server 1816 communicates with network 1808 via a communication link. In at least one embodiment, intranet server 1816 communicates with server manager 1818. In at least one embodiment, server manager 1818 includes an application configuration parameter database that is used by servers 1810, 1812, and 1814. In at least one embodiment, a user modifies database 1820 via intranet 1816, and server manager 1818 interacts with servers 1810, 1812, and 1814 to modify application parameters to match the contents of the database. In at least one embodiment, a user logs into intranet server 1816 by connecting to intranet 1816 via computer 1802 and entering authentication information (e.g., username and password).

[0155] In at least one embodiment, when a user wishes to register for a new service or modify an existing service, the intranet server 1816 authenticates the user and provides an interactive screen display / control panel that allows the user to access configuration parameters for a specific application. In at least one embodiment, multiple editable text boxes are presented to the user, describing configuration aspects of the user's website or other network resources. In at least one embodiment, if the user wishes to increase the storage space reserved for their website on the server, a field is provided for the user to specify the required storage space. In at least one embodiment, in response to receiving this information, the intranet server 1816 updates the database 1820. In at least one embodiment, the server manager 1818 forwards this information to the appropriate server and uses the new parameters during application runtime. In at least one embodiment, the intranet server 1816 is configured to provide the user with access to configuration parameters of hosted network resources (e.g., web pages, email, FTP sites, media sites, etc.) for which the user has a contract with a network hosting service provider.

[0156] Figure 19A A networked computer system 1900A is illustrated according to at least one embodiment. In at least one embodiment, the networked computer system 1900A includes multiple nodes or personal computers (“PCs”) 1902, 1918, and 1920. In at least one embodiment, the PC or node 1902 includes a processor 1914, memory 1916, a camera 1904, a microphone 1906, a mouse 1908, a speaker 1910, and a display 1912. In at least one embodiment, PCs 1902, 1918, and 1920 may each, for example, run one or more desktop servers on an internal network within a particular company, or may serve as servers for a general network not limited to a particular environment. In at least one embodiment, each PC node in the network corresponds to a server, thus each PC node in the network represents a specific network server and has a specific network URL address. In at least one embodiment, each server defaults to a default webpage for the user of that server, and this default webpage itself may contain embedded URLs pointing to other subpages on that user's server, or to other servers or webpages on other servers on the network.

[0157] In at least one embodiment, nodes 1902, 1918, 1920 and other nodes in the network are interconnected via medium 1922. In at least one embodiment, medium 1922 may be a communication channel, such as Integrated Services Digital Network (“ISDN”). In at least one embodiment, the individual nodes of the networked computer system may be connected via various communication media, including local area networks (“LANs”), ordinary telephone lines (“POTS”), sometimes also referred to as the Public Switched Telephone Network (“PSTN”) and / or variations thereof. In at least one embodiment, the individual nodes in the network may also constitute users of a computer system interconnected via a network such as the Internet. In at least one embodiment, each server on the network (operating from a specific node in the network at a given time) has a unique address or identifier in the network, which may be specified using a URL.

[0158] In at least one embodiment, multiple multipoint conferencing units (“MCUs”) can be used to transmit data to and from various nodes or “endpoints” of the conferencing system. In at least one embodiment, nodes and / or MCUs can be interconnected via ISDN links or local area networks (“LANs”), and also via various other communication media, such as nodes connected via the Internet. In at least one embodiment, nodes of the conferencing system can typically be directly connected to a communication medium such as a LAN, or they can be connected via MCUs, and the conferencing system can include other nodes or components, such as routers, servers, and / or variations thereof.

[0159] In at least one embodiment, processor 1914 is a general-purpose programmable processor. In at least one embodiment, the processor of a node in the networked computer system 1900A may also be a dedicated video processor. In at least one embodiment, the various peripheral devices and components of the node (e.g., the peripheral devices and components of node 1902) may differ from the peripheral devices and components of other nodes. In at least one embodiment, the configurations of nodes 1918 and 1920 may be the same as or different from node 1902. In at least one embodiment, the node may be implemented on any suitable computer system other than a PC system.

[0160] Figure 19BA networked computer system 1900B according to at least one embodiment is illustrated. In at least one embodiment, system 1900B illustrates a network (e.g., LAN 1924) that can be used to interconnect various nodes that can communicate with each other. In at least one embodiment, LAN 1924 has multiple nodes attached, such as PC nodes 1926, 1928, and 1930. In at least one embodiment, nodes may also be connected to the LAN via a network server or other means. In at least one embodiment, system 1900B includes other types of nodes or elements, such as routers, servers, and nodes.

[0161] In at least one embodiment, the neural network can be embedded in various devices, enabling these devices to predict future states and make proactive adjustments based at least in part on these predictions. In at least one embodiment, this can include directly mounting the neural network-enabled device, such as... Figure 20 The liquid-cooled server rack 2000 is shown. In at least one embodiment, the rack 2000 may include a plurality of liquid-cooled servers 2002 or other such devices. In at least one embodiment, the rack 2000 may include a rack manifold 2004 supporting a neural network for supplying a liquid flow to each liquid-cooled server 2002 via an inlet valve 2008 and returning the liquid with heat removed from the server 2002 via an outlet valve 2010. In at least one embodiment, the rack manifold 2004 may include a plurality of auxiliary liquid distribution control boards 2006 supporting a neural network. In at least one embodiment, one such control board 2006 may be provided for each inlet valve 2008 and outlet valve 2010 for each server. In at least one embodiment, the rack 2000 may also include a rack power distribution unit (PDU) 2012 supporting a neural network, which may include a plurality of ports designed to distribute power to computers or networked devices within the rack 2000. In at least one embodiment, each port of PDU 2012 can be associated with a power distribution control board 2014 supporting neural networks. In at least one embodiment, sensors can capture information about the computing environment inside and / or outside the rack 2000 (including inside and / or outside any individual server 2002 located therein), including temperature, airflow, or other such aspects. In at least one embodiment, liquid cooling can remove a certain amount of heat from the server 2002, but temperatures at different locations may vary due to factors such as load variations and external temperature fluctuations, and may reach or exceed temperature limits that allow these devices to continue operating normally. In at least one embodiment, efforts can be made to ensure that temperatures at specific locations remain below acceptable limits, where these locations may be related to the junction or core temperatures of processors (e.g., CPUs or GPUs), memory modules, or power supplies.

[0162] In at least one embodiment, the neural networks in these control boards 2006, 2014 can be responsible for predicting values ​​such as coolant flow rate or air flow rate, and the temperature of these various locations at one or more future points in time. In at least one embodiment, the corresponding boards, manifolds, PDUs, or other such devices can be adjusted using the predictions from these networks to ensure appropriate temperature and flow rates and to ensure proper device operation. In at least one embodiment, these adjustments can include adjusting the amount of heat transfer fluid or air flowing into or out of the server or device, and the temperature of that fluid or air. In at least one embodiment, this can also include adjusting the amount of power supplied to one or more of these devices. In at least one embodiment, these adjustments can include adjustments similar to those made in other data centers, but these adjustments can be made based on future predictions rather than observed states, thereby enabling the system to operate proactively rather than reactively. In at least one embodiment, such monitoring can be continuous, or at least periodically, to ensure continuous and normal operation.

[0163] In at least one embodiment, each neural network in each component can be specifically trained for that component type and can receive updated network parameters generated as a result of further training or continuous learning. In at least one embodiment, inference results generated by individual devices can also be shared with other devices in the data center, which helps to make more accurate predictions. In at least one embodiment, each control board 2006, 2014 can be a neural network-enabled board, ASIC, or other such component. In at least one embodiment, the control board can include components of a general-purpose computer, including at least one processor and memory, and associated circuitry. In at least one embodiment, a Jetson board from NVIDIA, which is a complete system module (SPM) containing a CPU, GPU, PMIC, DRAM, and flash memory, can be used. In at least one embodiment, such modules can also be expanded to provide additional functionality or capabilities. In at least one embodiment, each such board can operate as a small, AI-enabled or machine learning-enabled computer. In at least one embodiment, multiple such boards can be used in parallel, for example on rack manifold 2004, to process data from multiple high-resolution sensors concurrently. In at least one embodiment, these different neural networks can make predictions that can be shared between servers, racks, or data centers, for example, to provide accurate predictions for various locations at one or more future points in time. In at least one embodiment, appropriate adjustments can then be made as needed to maintain temperature and other parameters at appropriate levels or values. In at least one embodiment, the liquid distribution control board 2006 can make predictions that can be used to adjust control only on relevant flow values, or these predictions can be shared to allow for further adjustments based at least in part on these predictions. In at least one embodiment, the control system for the data center can collect predictions from these different boards or networks to allow for adjustments more suitable for the entire data center. In at least one embodiment, adjustments can be made at the server, rack, pod, data center, or other similar level.

[0164] In at least one embodiment, a data center-class rack liquid cooling method enabled by neural networks can utilize... Figure 21AThe components of the system 2100 shown. In at least one embodiment, the external cooling unit 2102 can provide liquid at a determined temperature to the data center cooling distribution unit 2104. In at least one embodiment, the unit 2104 may include a set of neural network-enabled distribution unit control boards 2106 that can perform inference and adjust based on inferred future values. In at least one embodiment, this may include adjusting the temperature of the liquid to be received from the external cooling unit 2102, or adjusting the flow rate of the liquid into and / or out of the external unit 2102. In at least one embodiment, this may also include adjusting the flow rate into manifolds 2108 for a row of liquid-cooled racks 2112, and the flow rates into and out of these racks, as can be determined at least in part based on the inference results of the neural network-enabled control boards 2110 built into the manifolds 2110. In at least one embodiment, the flow rate levels into and out of different liquid-cooled racks may be different. In at least one embodiment, the flow rate into each server in the rack may also be different, as combined with Figure 20 As described. In at least one embodiment, a similar method can also be used in air-cooled data centers. In at least one embodiment, temperature and airflow can be measured, and future values ​​can be predicted, for example, over the next 30 seconds, 1 minute, 5 minutes, or 1 hour, thereby adjusting air temperature, airflow, power distribution, and other relevant aspects. In at least one embodiment, these predictions can be made using various types of input data, such as server load, current temperature, current pressure, flow rate, power consumption, and other relevant data from different locations within the data center.

[0165] In at least one embodiment, these neural network-enabled boards can control the flow, supply, and return of liquids at different locations (e.g., racks, servers, or GPU levels). In at least one embodiment, these boards can be used to predict, control, and create optimal solutions for data centers to maintain required operating conditions and avoid equipment loss or downtime due to temperature issues or thermal events. In at least one embodiment, this can be based on current and anticipated loads and measured or detected network conditions. In at least one embodiment, this approach can also be applied to other types of devices, such as UPS power supplies. In at least one embodiment, any device with logic boards or components can be injected with such small neural network code to achieve the above and other similar purposes. In at least one embodiment, such a network can access all available devices in a unit or device, such as in devices like PDUs, where boards can adjust or completely cut off power to server nodes, for example, in the event of a detected or predicted leak and potential damage to the associated equipment if continued operation is possible. In at least one embodiment, operation can be gracefully terminated or power can be shut off to avoid damage. In at least one embodiment, an executing program can be terminated to at least prevent data loss.

[0166] In at least one embodiment, the cooling system may employ a direct recirculation design, such as... Figure 21A As shown. In at least one embodiment, coolant from CDU 2104 is directed along supply manifold 2108 to the individual server racks. In at least one embodiment, the first server rack (A) will receive liquid first along this flow, while the last server rack (E) will receive liquid last along this flow. In at least one embodiment, return manifold 2110 will be used to receive the heated liquid passing through these racks and return the heated liquid to CDU 2104. In at least one embodiment, the heated liquid returning from the first server rack (A) has the shortest return path, while the heated liquid returning from the last server rack (E) has the longest return path. In at least one embodiment, this results in the overall liquid path 2114 of the first server rack (A) being much shorter than the overall liquid path 2116 of the last server rack (E) 2116. In at least one embodiment, factors such as pressure and flow rate may vary significantly throughout the cooling system loop. In at least one embodiment, this may include lower pressure near the ends of the cooling loop and higher pressure in the middle of the loop. In at least one embodiment, this unbalanced design can lead to unacceptable variations in flow rates in data centers or other environments because different path lengths result in varying pressures, temperatures, and other environmental factors that affect these flow rates. In at least one embodiment, the different amounts of time required for the liquid to flow along these different paths also contribute to these flow rate variations.

[0167] In at least one embodiment, the following can be employed: Figure 21B The system 2120 is based on a reverse flow design. In at least one embodiment, the system includes a reverse flow manifold 2122 that first receives heated liquid from a first server rack (A) and finally receives heated liquid from a last server rack (E), but instead of directly guiding the flow back to CDU 2106, it allows the liquid to flow from the first server rack to the last server rack (e). Figure 21B As shown (from right to left), the fluid for each server rack will flow at least through the corresponding manifold along a path 2124 of similar length. In at least one embodiment, these path lengths may be virtually equal, allowing for slight variations depending on factors such as layout and component type. In at least one embodiment, the fluid for each server rack will flow through almost the entire length of the supply manifold and the return manifold, rather than as shown... Figure 21A Unlike a direct return design, the liquid flows through different sections. In at least one embodiment, this ensures that the liquid in all server racks experiences the same variations in temperature, pressure, and other environmental parameters throughout the manifold's operation. In at least one embodiment, this balanced design helps reduce flow rate variations of the liquid delivered to different locations throughout the cooling system. In at least one embodiment, this helps ensure that flow rate variations remain within the desired range or below a maximum variation threshold. In at least one embodiment, this balanced design helps ensure that the individual cooling components of the liquid cooling system deliver equal and sufficient amounts of liquid to all servers and racks in the data center.

[0168] In at least one embodiment, all supply and return fluid lines between one or more CDUs and one or more direct-to-chip (D2C) cold plate cooling loops or heat exchangers in immersion liquid-cooled blade servers can employ reverse recirculation, rather than direct recirculation, in various data center manifolds, exhaust manifolds, and rack manifolds. In at least one embodiment, electronic valve controls at each outlet of these manifolds, exhaust manifolds, and rack manifolds are capable of monitoring and controlling the flow rates of primary and / or auxiliary heat transfer fluids into and out of these CDUs, which lead to cold plates attached to components such as CPUs, GPUs, and switches, as well as other D2C liquid-cooled components and immersion blade servers. In at least one embodiment, the liquid cooling system can employ a self-balancing flow design in all primary and auxiliary loops, combined with electronic monitoring and control utilizing AI or machine learning, based at least in part on data received from various flow rate, pressure, temperature, power, and other environmental or operational sensors. In at least one embodiment, this approach can improve or optimize liquid distribution to liquid-cooled components such as servers and processors in environments such as data centers.

[0169] In at least one embodiment, neural network-based flow control can be used with this design to provide further variation minimization or flow consistency. In at least one embodiment, the combination of self-balancing with electronic monitoring and control can provide a highly balanced and optimized coolant flow for each of servers, racks, processors, or other such components, without these components competing with each other or requiring complex control schemes. In at least one embodiment, flow controllers can be located at various points throughout the cooling system, for example, to guide coolant inflow and outflow from components such as racks, servers, and electronic components, where the liquid can be supplied via paths such as manifolds and rack manifolds. In at least one embodiment, these flow controllers can implement incremental control (e.g., partial opening and closing) to provide an accurate self-balancing approach. In at least one embodiment, a self-balancing design with finely adjustable control can provide accurate balancing of liquid flow at each (if not all) outlet of the liquid cooling system. In at least one embodiment, this method can be used to provide, for example, a flow rate of 50 liters per minute (lpm) with a tolerance, threshold, or range of variation of + / - 5 lpm. In contrast, a design without this method may have a flow rate of approximately 60 lpm at one location and approximately 40 lpm at another location due to differences such as pressure gradients and gravity effects, which could be an unacceptable flow imbalance in the entire system.

[0170] In at least one embodiment, the control system 2140 can be as follows: Figure 21CThe implementation is shown. In at least one embodiment, a reverse-flow manifold design can be used to provide pressure and flow balance throughout the system. In at least one embodiment, a controllable flow valve 2146 can be used at various inlet and / or outlet locations, such as inlets and outlets from the manifold to individual racks. In at least one embodiment, the flow manager 2142 can receive data from various sensors, devices, or data sources in the relevant environment. In at least one embodiment, the flow manager 2142 can provide at least a portion of this data as input to one or more neural networks (e.g., deep neural networks (DNNs)) of the inference module 2144. In at least one embodiment, the sensors and valves can be calibrated for the type of heat transfer fluid used. In at least one embodiment, there can be one flow manager 2142 for the entire data center or environment, or multiple flow managers can be set for individual rows, racks, or servers. In at least one embodiment, neural networks can also be used at each individual flow control valve, such as those communicating with one or more independent flow controllers. In at least one embodiment, the flow manager 2142 can receive one or more inference results from the inference module 2144, wherein these inference results can include adjustments or settings for the individual flow control valves. In at least one embodiment, the flow manager 2142 may send this data or related data or instructions to these individual flow control valves 2146 or other such components for minor adjustments to help balance the flow across the entire system.

[0171] In at least one embodiment, this method can be used at the rack level or server level, such as... Figure 21D The system 2160 is shown in the diagram. In at least one embodiment, the liquid-cooled rack may have a rack manifold with a reverse reflux design, thereby ensuring that the liquid path lengths for each server in the rack are equal. In at least one embodiment, the size of the manifold may be much smaller than the size of the corresponding rack, such as... Figure 21D The exploded view shown is provided for clarity. In at least one embodiment, the flow manager 2162 may utilize the inference module 2164 to determine the necessary adjustments to the various flow control valves 2166 within the rack to balance flow between the servers in the rack. In at least one embodiment, the flow manager may also provide such instructions to the flow control valves or components within the servers, for example, to direct coolant to specific cold plates, processors, or other internal server components. In at least one embodiment, using these intelligent control valves can provide well-balanced piping distribution and controllability. In at least one embodiment, each of these intelligent control valves 2166 can be fine-tuned as needed, such as under the guidance of the flow manager. In at least one embodiment, these flow valves may employ PID or PIV servo control, where PIV allows control based on both position and velocity errors.

[0172] In at least one embodiment, one or more neural networks can be trained to infer the amount of liquid cooling required for various components under various conditions. In at least one embodiment, this may include determining how much additional cooling might be needed when the GPU's operating power increases from a more typical 300 watts to 500 watts, which might require 1.7 lpm instead of 1.0 lpm. In at least one embodiment, the neural network can infer adjustments to the liquid flow rate of the component and can enable proactive and accurate adjustments to avoid significant temperature fluctuations. In at least one embodiment, one or more operation tables may exist that correlate factors such as proportional flow rate with power consumption; these operation tables can be used to train these models or to determine adjustments based on inferred state data. In at least one embodiment, these tables can also be used for different types of components or different models of the same type of component to account for the inherent differences between these components. In at least one embodiment, inference can be performed at different levels of granularity, from the entire data center level to the level of internal server components (e.g., GPUs). In at least one embodiment, data may be collected by a component such as a board-level management controller (BMC) or a server and provided to a flow manager via a network port; the flow manager can send instructions to one or more flow controllers to make one or more appropriate adjustments.

[0173] In at least one embodiment, the neural network can predict the fluid state at various locations at one or more future times, at least in part, based on available information. In at least one embodiment, this information may include snapshots of current information or may include at least some recent data. In at least one embodiment, these predictions can be used to adjust aspects such as flow rate. In at least one embodiment, the cooling distribution unit can adjust the liquid flow rate or coolant temperature. In at least one embodiment, such devices can utilize these predictions and proactively take measures to prevent any adverse events before they occur. In at least one embodiment, various sensors can provide information about the current state of the computing environment. In at least one embodiment, this may include the use of sensors such as temperature sensors, load sensors, flow sensors, or pressure sensors, whose data may be acquired instantaneously or historically. In at least one embodiment, predictions based on data from these sensors can be compared with one or more thresholds, ranges, or other operating criteria to determine whether any changes are needed. In at least one embodiment, this may include adjustments to prevent unacceptable temperature rises at specific locations in a data center or other similar environment. In at least one embodiment, this may include remedial measures such as closing valves or increasing coolant flow, adjusting coolant temperature, or issuing alarms.

[0174] The term "coupled" or similar expressions are intended to encompass both direct and indirect connections, whether physical, electrical, mechanical, logical, or otherwise. A connection can involve direct physical contact, attachment, or engagement, or it can involve communication, interaction, or collaboration between elements. The term "coupled to" is used broadly to cover any form of association, link, or relationship that enables the desired functionality or collaboration between two declared elements. The term "coupled to" is not limited to any particular type of connection and is intended to cover a range of possible configurations and implementations.

[0175] Although the operations of the methods herein are shown and described in a specific order, the order of operations for each method can be changed so that some operations can be performed in reverse order, or some operations can be performed at least partially concurrently with other operations. In another embodiment, instructions or sub-operations of different operations can be implemented in an intermittent and / or alternating manner.

[0176] Alternatively, embodiments of this disclosure can be implemented entirely in hardware, or with an implementation scheme that includes both hardware and software elements. In embodiments using software, the software may include, but is not limited to, firmware, resident software, microcode, etc.

[0177] While specific embodiments of this disclosure have been described and illustrated, this disclosure is not limited to the specific forms or arrangements of components so described and illustrated. The scope of this disclosure is defined by the appended claims and their equivalents.

Claims

1. A connector, comprising: A fixed interface component is coupled to the main structure, such that the fixed interface component is fixed relative to the main structure; as well as A piston assembly operably coupled to the body structure and operably coupled to the fixed interface member through the body structure, the piston assembly including a piston connected to the coupling member, the piston being located in the piston assembly and configured to be axially and radially displaced relative to the body structure, such that the coupling member is movable along the axial and radial directions.

2. The connector as claimed in claim 1, wherein, The coupling component is a universal quick disconnect device (UQD).

3. The connector of claim 1 further includes a friction-reducing washer located between the piston assembly and the body structure, allowing the piston assembly to move along the radial direction.

4. The connector as claimed in claim 1, wherein, The elongated portion of the piston is located in a piston bore member operatively coupled to the main body structure, wherein the piston is movable relative to the piston bore member along the axial direction.

5. The connector of claim 4, further comprising an O-ring located on the elongated portion of the piston, the O-ring providing a seal between the elongated portion of the piston and the piston bore member.

6. The connector of claim 4, further comprising an O-ring located between the fixed interface member and the piston bore member, the O-ring providing a seal between the fixed interface member and the piston bore member.

7. The connector of claim 4 further includes a helical spring located between the piston and the piston bore member, the helical spring being used to return the piston to a default position after axial movement.

8. The connector of claim 7, further comprising a friction-reducing washer located between the helical spring and the piston to enable the piston to move along the radial direction.

9. The connector as claimed in claim 1, wherein, The fixed interface component is a closed barb with a barb assembly end, which is used to connect to the fluid pipeline.

10. A connector for connecting a rack and a switch, comprising: A fixed interface component is coupled to the main structure such that the fixed interface component is fixed relative to the main structure, and the fixed interface component is connected to the switch; A piston assembly operably coupled to the body structure and operably coupled to the fixed interface member through the body structure, the piston assembly including a piston connected to a coupling member connected to the frame, the piston being located in the piston assembly and configured to be axially and radially displaced relative to the body structure, thereby allowing axial and radial floating on the frame side of the connector while remaining fixed on the switch side of the connector.

11. The connector of claim 10, wherein, The coupling component is a universal quick disconnect device (UQD) connected to the rack.

12. The connector of claim 10, further comprising a friction-reducing washer located between the piston assembly and the body structure, allowing the piston assembly to move along the radial direction.

13. The connector of claim 10, wherein, The elongated portion of the piston is located in a piston bore member operatively coupled to the main body structure, wherein the piston is movable relative to the piston bore member along the axial direction.

14. The connector of claim 13, further comprising a helical spring located between the piston and the piston bore member, the helical spring being used to return the piston to a default position after axial movement.

15. The connector of claim 14, further comprising a friction-reducing washer located between the helical spring and the piston to enable the piston to move in the radial direction.

16. The connector of claim 10, wherein, The fixed interface component is a closed barb with a barb assembly end, which is used to connect to the fluid pipeline in the switch.

17. A liquid-cooled exchanger, comprising: case; The housing contains piping for transmitting coolant; as well as A connector with two degrees of freedom is configured to counteract the dependence of movement inside and outside the switch, such that when a portion of the connector moves outside the switch, the conduit in the switch connected to the connector does not move.

18. The liquid-cooled exchange as described in claim 17, wherein, The connector includes a coupling member that is movable in both the axial and radial directions.

19. The liquid-cooled exchange as described in claim 18, wherein, The connector includes an interface member that remains stationary as the coupling member moves in the axial and / or radial direction, and the interface member is connected to the conduit.

20. The liquid-cooled exchange as described in claim 17, wherein, The connector includes a piston and a piston bore member, wherein the piston is configured to move axially relative to the piston bore member.

21. The liquid-cooled exchange as described in claim 17, wherein, The pipeline can be a rigid pipeline, a flexible pipeline, a corrugated pipeline, or a welded pipeline.

Images