Segmented architecture for fluid management and leak sensors

Abstract

A fluid manifold can be assembled into an information technology (IT) rack to deliver and distribute fluids to IT equipment. The manifold may include multiple sections, each with one or more shut-off valves. One or more leak detection sensors can be arranged to detect leaks in any section and in any IT equipment. A controller can control the shut-off valves to the closed position based on detected leaks. This design enables the manifold to better manage and control fluids used in mission-critical IT equipment.

Description

Technical Field

[0001] Embodiments of this disclosure generally relate to an information technology (IT) rack. In particular, the IT rack may have a segmented architecture for fluid management and leak sensors. Background Technology

[0002] Information technology (IT) includes technologies such as computers accessible via the internet or local area networks, which provide storage or access to data, websites, computer programs, algorithms, services, etc. IT equipment, such as servers and other electronic devices (e.g., peripherals), can be installed in chassis. These chassis can then be installed in IT racks. An IT rack can contain multiple chassis in a way that manages the power and heat requirements of the IT equipment.

[0003] Liquid cooling systems transfer and deliver liquid between IT racks and chassis, and between chassis and the IT equipment mounted on them. Such systems provide high operational and cooling efficiency for IT equipment. Liquid cooling is becoming increasingly popular for high-power-density electronic devices because air cooling may be insufficient in some situations.

[0004] Liquid cooling systems are prone to fluid leaks, which can damage IT equipment. When a leak is detected in an IT rack, the entire rack may be shut down to minimize damage to the equipment; however, this can result in unintended capacity loss and reduced data center reliability. Furthermore, cooling systems can lack flexibility in how components can be arranged or rearranged within IT racks. Permanent, monolithic cooling solutions make it difficult to have different arrangements of IT equipment across racks and to rearrange IT equipment within racks. Therefore, a robust liquid cooling architecture is needed that resists leak failures and provides flexible operation of IT equipment even when a leak is detected. Summary of the Invention

[0005] The present invention aims to solve one of the aforementioned technical problems in the related art.

[0006] This invention provides a liquid cooling system for an information technology (IT) rack, comprising: a liquid manifold including multiple sections, each of the multiple sections having one or more shut-off valves; one or more leak detection sensors; and a controller having one or more processors configured to: receive the status of the one or more leak detection sensors; and, in response to a first leak detection sensor among the one or more leak detection sensors detecting a leak, control a first shut-off valve among the one or more shut-off valves associated with the first leak detection sensor to a closed position.

[0007] In some embodiments, each of the plurality of sections delivers cooling fluid to a corresponding IT device that is vertically stacked together on the IT rack.

[0008] In some embodiments, in the IT rack, the IT equipment of higher importance is positioned above the IT equipment of lower importance.

[0009] In some embodiments, the processor is configured to control a second shut-off valve among the one or more shut-off valves to open when a second leak detection sensor among the one or more leak detection sensors does not detect a leak. The second shut-off valve is arranged to control flow to a second IT device positioned above a first IT device among the IT devices, which is shut off from receiving fluid by the first shut-off valve among the one or more shut-off valves in the closed position.

[0010] In some embodiments, the processor is further configured to communicate with the first IT device and the second IT device in the IT device in response to the leak, so as to migrate one or more operations performed by the first IT device in the IT device to the second IT device in the IT device.

[0011] In some embodiments, each of the plurality of segments includes a first connector at a first end and a second connector at a second end, wherein the first connector and the second connector are blind-fit connectors or quick-disconnect connectors.

[0012] In some embodiments, at least one of the first connector and the second connector includes a corresponding shut-off valve among the one or more shut-off valves.

[0013] In some embodiments, both the first connector and the second connector include a corresponding shut-off valve from the one or more shut-off valves.

[0014] In some embodiments, the one or more leak detection sensors are arranged in each of the plurality of segments and in each corresponding IT device to detect whether any leak exists in any of the plurality of segments or in any of the corresponding IT devices.

[0015] In some embodiments, the liquid cooling system further includes a second manifold arranged to deliver cooling fluid to IT equipment in the IT rack in response to the leak.

[0016] This invention also provides an information technology (IT) rack, including a liquid cooling system according to an embodiment of the invention. Attached Figure Description

[0017] In the accompanying drawings, various aspects are illustrated by way of example rather than limitation. The same reference numerals in the drawings indicate similar elements. It should be noted that references to “a” or “an” aspect of this disclosure do not necessarily refer to the same aspect, and they mean at least one. Moreover, for the sake of brevity and to reduce the total number of drawings, a given drawing may be used to illustrate features of more than one aspect, and not all elements in the drawing are required for a given aspect.

[0018] Figure 1 A cooling system with a segmented architecture according to some embodiments is shown.

[0019] Figure 2 An IT rack with a cooling system is shown according to some embodiments.

[0020] Figure 3 Modular sections of a manifold according to some implementations are shown.

[0021] Figure 4 A modular manifold according to some implementations is shown.

[0022] Figure 5 and Figure 6 An example of a modular section of a manifold with integrated valves and connectors, according to some embodiments, is shown.

[0023] Figure 7 An IT rack with a cooling system having redundant manifolds is shown according to some embodiments.

[0024] Figure 8 This is a block diagram illustrating an example of an IT rack with an integrated cooling system. Detailed Implementation

[0025] Several aspects of this disclosure will now be explained with reference to the accompanying drawings. Unless the shape, relative positions, and other aspects of the parts described in a given aspect are explicitly defined, the scope of this disclosure is not limited to the parts shown, which are for illustrative purposes only. Furthermore, while many details are set forth, it should be understood that some aspects can be practiced without these details. In other instances, known circuits, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification. Moreover, unless the meaning is clearly to the contrary, all scopes set forth herein are to be considered to include the endpoints of the respective scopes.

[0026] The reference to "one embodiment" or "implementation" in this specification means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this disclosure. The phrase "in one embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0027] In some aspects of this disclosure, a liquid cooling system for an IT rack is described. The liquid cooling system may have a liquid manifold having multiple sections. Each section may have one or more shut-off valves that can cut off fluid to all IT equipment receiving fluid from that section. One or more leak detection sensors may be arranged to sense leaks in the manifold and / or the IT equipment. A controller may receive the status of one or more leak detection sensors. If a leak is detected, the controller may control the first of one or more shut-off valves associated with the first of the one or more leak detection sensors to a closed position. For example, if a leak is detected from a first IT equipment or from a section of the manifold, the controller may control the first shut-off valve, which is immediately upstream of the IT equipment and associated with that section, to a closed position. This solution provides rack-level design for implementing advanced liquid management and fluid control, as well as leak detection sensors.

[0028] Figure 1 A cooling system 100 with a segmented architecture according to some embodiments is shown. The cooling system may have a fluid inlet line 102, which can be connected to a fluid source to receive fluid, such as a liquid coolant (e.g., water, propylene glycol, etc.). A main valve 104 may be arranged in the path between the fluid inlet line and the fluid manifold 106. If the main valve is closed, the fluid to the entire manifold 201 is cut off. If the main valve is open, fluid can be delivered from the fluid source to the manifold.

[0029] Liquid manifold 106 may include multiple sections, such as sections 120, 121, 122, and 123. A liquid manifold (e.g., liquid manifold 106) that supplies and distributes fluid to IT equipment may be referred to as a supply manifold. A manifold that receives fluid from IT equipment may be referred to as a return manifold. Each of the multiple sections may have one or more shut-off valves. Each valve can control whether fluid is delivered to the corresponding section and downstream sections or is shut off. For example, section 121 may have valve 131. If valve 131 is open, fluid is delivered to section 121 and downstream sections of section 121, such as sections 122 and 123. If valve 131 is closed, section 121 and downstream sections 122 and 123 are disconnected from the fluid source. Section 122 also has a fluid valve 132, which may also be controlled to a closed position, etc. It should be noted that although section 120 is not shown as having a valve in the body of the section, the main valve 104 can be considered by the controller to belong to section 120. Thus, if a leak exists in the upstream section, the main valve can be controlled to the closed position. In some embodiments, as shown in other figures, the upstream section (e.g., section 120 in this example) may also have a valve.

[0030] Each segment may have one or more fluid ports 108. IT equipment can connect to each port to receive fluid from the corresponding segment. Different segments may have the same or different numbers of ports. Similarly, segments may be of uniform length or have varying lengths. Segments may be connected in series to form a single flow path from the upstream segment to the downstream segment. Fluid ports can be understood as fluid connectors, such as manual-mating drip-free connectors or blind-mating drip-free connectors.

[0031] The cooling system may include one or more leak detection sensors 151, 152, 153, and 154. For example, the system may include at least one leak detection sensor that can be associated with a corresponding fluid valve 132. The leak detection sensor may include nanoparticle sensors, chemical sensors, sensing bands, and / or other equivalent sensing technologies that sense the presence of fluid. Alternatively or additionally, the leak detection sensor may sense fluid flow or pressure at two or more locations to determine the presence of a leak. Each leak detection sensor may be associated with a valve immediately upstream of the leak monitored by the leak detection sensor. For example, each leak sensor 154 may be positioned on or on an IT device to which fluid is delivered by segment 123. Since there are no other fluid valves between fluid valve 133 and segment 123, fluid valve 133 may be associated with each leak sensor 154. This association may also operate in the other direction. For example, each valve may be associated with sensors arranged to detect leaks from a segment immediately downstream of the valve. For example, one or more sensors 151 may be positioned to detect leaks from manifold section 120 and / or from IT equipment receiving fluid from section 120. Similarly, sensor 152 may be positioned to detect leaks in section 121 of the manifold and / or from IT equipment receiving fluid from section 121.

[0032] Controller 122 can receive the status of one or more leak detection sensors and, in response to a leak being detected by any of the leak detection sensors, control one of the shut-off valves associated with the detected leak to the closed position. The sensors can be hardwired to the controller and / or communicate with the controller wirelessly. The controller can access information that maps each sensor to each shut-off valve directly upstream of it. This information can be stored in a computer-readable component as a table or other data structure. Thus, when the controller receives a status indicating that sensor 152 has detected a leak, the controller can control the fluid valve directly upstream (e.g., fluid valve 131) to the closed position. Other valves (e.g., upstream valves) can be controlled to remain open. Fluid flows downstream from the fluid inlet, sequentially through each segment, thus the closed upstream valve cuts off fluid flow to all IT equipment downstream of the valve.

[0033] Operations, methods, and processes performed by the controller can be executed using processing logic 123. Processing logic may include hardware (e.g., circuitry, special-purpose logic, etc.), software (e.g., implemented on a non-transitory computer-readable medium), or a combination of both. Although the processes or methods described above are executed in a certain order, it should be understood that some of the operations described may be performed in a different order. Furthermore, some operations may be performed in parallel rather than sequentially.

[0034] Figure 2 An IT rack 200 with a cooling system according to some embodiments is shown. The cooling system may include features described in other parts of this disclosure. Manifold 201 may have multiple sections, such as sections 202 and 204. Each of the multiple sections delivers cooling fluid to a corresponding IT device 210 that is vertically stacked together on the IT rack. The IT device may have flexible or rigid conduits attached to each port to receive the cooling fluid. Section 202 may deliver fluid to IT devices A, B, C, D, and E through corresponding fluid ports of section 202. The cooling fluid passes through the IT devices, thereby absorbing heat energy from the IT devices, and may exit the cooling system and IT rack via manifold return section 203. The fluid may be cooled by facility cooling units and systems (e.g., chillers, refrigeration, or other cooling technologies) and recirculated to the IT rack or other IT racks.

[0035] One or more leak detection sensors (241 and 242) can be deployed in each of the multiple sections and in each corresponding IT device to detect the presence of any leaks in any of the multiple sections or in any of the corresponding IT devices. For example, one or more leak sensors 241 can be strategically deployed on, near, or below any of the IT devices AE to determine whether any of the IT devices AE has a fluid leak. Alternatively or additionally, leak sensors 241 can be strategically positioned on, near, or below a section 202 of a manifold to detect leaks in that section of the manifold or in the piping from the manifold to the IT devices. Leak sensors can also be integrated with the packages of individual servers; a leak sensor can be understood as a set of sensors integrated with multiple servers.

[0036] Similarly, leak sensor 242 may be positioned on, near, and / or below any one or each of the IT equipment FJ to detect whether any of them is leaking. Alternatively or additionally, leak sensor 242 may be positioned on, near, and / or below section 204 to detect the presence of fluid leakage from section 204 of the manifold. In some aspects, leak sensor 242 may be understood as a sensor integrated within one or more sensors within a server or within a cooling module.

[0037] As shown in the figure, IT devices can be stacked one on top of another in the various shelves of IT rack 200. Therefore, a leak from section 202 of IT device AE or manifold could travel downwards to IT device FJ and cause damage. However, due to gravity, a leak from device FJ or section 204 is less likely to travel upwards and cause damage to IT device AE. This gravity frame can be used to implement a priority ordering of IT devices.

[0038] In an IT rack, IT devices of higher importance (e.g., running more critical workloads or requiring more critical Service Level Agreements (SLAs)) can be positioned above IT devices of lower importance. For example, IT device A can perform critical operations. Therefore, IT device A can be grouped with other IT devices (e.g., IT devices B, C, D, and E) that receive fluid from manifold 201 (the upstream manifold in this example). However, IT device H can be considered to perform non-critical operations, and thus, it is grouped with other IT devices (F, H, I, and J) considered to be of lower importance.

[0039] The group (IT equipment FJ) receives fluid from section 204. If a leak is detected in one or more leak detection sensors 242 located downstream, the controller can close the shut-off valve 234, which is the immediate upstream valve, to reduce the risk of damage to the IT equipment in the group (e.g., IT equipment FJ). Because the IT equipment is stacked vertically in this example (e.g., relative to gravity), a leak detected in sensor 242 poses minimal risk to IT equipment AE.

[0040] Therefore, the controller can control a first shut-off valve (e.g., valve 234) immediately upstream of the detected leak to the closed position. An upstream second shut-off valve (e.g., main valve 232) can be controlled to open in response to an upstream leak detection sensor (e.g., leak detection sensor 241) not detecting a leak. The second shut-off valve (e.g., valve 232) controls flow to a second IT device (e.g., IT device AE), positioned above a first IT device (e.g., IT device FJ), which is arranged to receive fluid through the first shut-off valve (e.g., valve 234). More broadly, all valves except the second shut-off valve (e.g., any valve upstream of the second shut-off valve) can be controlled to remain open.

[0041] The controller can manage the shutdown of each IT device when the valve upstream of each IT device is closed. For example, if shut-off valve 234 is closed, the controller can manage the shutdown of IT device FJ by interrupting its power supply. The controller can keep IT device AE powered. However, if the controller closes valve 232, the controller can interrupt the power supply to both IT devices AE and FJ.

[0042] In some implementations, in response to the detection of a leak, the controller may also communicate with a first IT device (e.g., IT device F) and a second IT device (e.g., IT device E) to migrate one or more operations performed by the first IT device to the second IT device.

[0043] For example, if IT device F includes a server that performs movie streaming operations, the controller can assign these movie streaming operations to IT device E for execution. This can include adjusting the software, settings, and / or firmware of IT device E. By migrating operations in response to a leak, the controller can mitigate the impact of the leak and improve the overall reliability of the IT equipment.

[0044] The controller can communicate with each of the IT devices, valves, and leak detection sensors via one or more wired or wireless communication channels. These communication channels may include network communication protocols (such as TCP / IP), analog I / O signals, digital I / O signals, and / or digital communication buses. The valve may have an actuator, such as a spring-loaded solenoid, which is normally open unless commanded to close by the controller, servo motor, or other controllable actuating mechanism.

[0045] Figure 3 A modular section 300 of a manifold according to some embodiments is shown. The modular section may include a first connector 302 at a first end and a second connector 304 at a second end. A fluid shut-off valve 308 may be disposed upstream of each fluid port 306 in the section. The fluid ports may be drip-free quick-disconnect. In some embodiments, the first and / or second connectors may include blind-mating or quick-disconnect connectors. Blind-mating and quick-disconnect connectors may have tabs, clips, washers, springs, and / or other combinations of mechanisms that provide sliding, twisting, and / or snapping of mating connectors in a fluid-tight connection. Such connectors offer improved maintainability and modularity, meaning the section can be manually assembled and disassembled without tools (e.g., wrenches, welding, cutting, etc.).

[0046] Figure 4 A modular manifold 400 according to some embodiments is shown. The manifold may include multiple modular sections 402, 404, and 406. These modular sections may be as shown in other figures (e.g.) Figure 3 , Figure 5 or Figure 6The sections described are as follows. Each section may have a corresponding fluid shut-off valve, located at the upstream end of the corresponding section and upstream of its corresponding fluid port within that section. The downstream end of one section connects to the upstream end of another section. Therefore, the management of fluid cooling for IT equipment can be flexible. If the layout of IT equipment changes, or if there are critical changes to the IT equipment within the rack, the manifold can be modified relatively easily accordingly. It should be noted that some modular sections may have different lengths or numbers of ports than other sections. For example, section 406 may be longer and have more ports than sections 402 and 404. Sections 402 and 404 may have the same length and number of ports. These sections are mixed and combined to suit the specific layout of IT equipment within the IT rack.

[0047] Figure 5 and Figure 6 An example of a modular section of a manifold with integrated valves and connectors, according to some embodiments, is shown. Figure 5 The image shows a modular section 500 with a first connector 501 located at the upstream end of the modular section. This first connector may have a shut-off valve integrated or built into the connector. This simplifies the design of the modular section 500, reduces the number of connectors and adapters needed to accommodate shut-off valves in the pipeline, and thus reduces potential points of failure. In some embodiments, such as... Figure 6 As shown, the two end connectors 601 and 602 of the modular section 600 may have integrated built-in shut-off valves. This feature simplifies the installation of the modular section and reduces the risk of incorrect orientation.

[0048] Figure 7An IT rack 700 with a cooling system having redundant manifolds is shown according to some embodiments. As discussed, the cooling system may include a manifold 701 comprising multiple sections, which may selectively cut off fluid to IT equipment in some sections based on where a leak is detected. In some cases, if a leak is detected in a section of manifold 701, the controller may switch operation to a backup manifold 702. As shown in this example, the backup manifold may have a single shut-off valve that opens when a leak is detected in manifold 701. Assuming relatively infrequent leaks, the backup manifold may supply fluid to the IT equipment until manifold 701 or the leaking section of the manifold is repaired or replaced. In some embodiments, the backup manifold is used when the upstream valve of the main manifold is closed. Otherwise, the main manifold may operate in other sections as discussed, some of which are active while other upstream IT equipment is shut down due to a detected leak. In some embodiments, the backup manifold 702 covers other scenarios, such as when a leak occurs in the main manifold 702. Therefore, if a leak sensor installed inside the server detects a leak in the server, the backup manifold may not activate because, in this case, the fluid supply to the server needs to be shut off. For example, if a leak sensor inside IT device H is triggered, the valve at the bottom of the main manifold 701 can be closed. The backup manifold 702 will also remain inactive (e.g., the fluid shut-off valve of the backup manifold remains in the closed position). IT device GJ will be shut down. Otherwise, if the backup manifold 702 is active, IT device H will continue to leak. On the other hand, if the sensor detects a leak from manifold 701 but not from the IT device, the controller can shut off manifold 701 and activate the backup manifold 702. In this case, IT device GJ can remain active.

[0049] In other embodiments, the backup manifold 702 may be segmented (e.g., having one or more controllable downstream fluid shut-off valves). In this case, the backup manifold can be arranged identically to the main manifold, such that if a section leak is detected, that section (and its downstream sections) can be shut off. The backup manifold can seamlessly direct fluid to these sections without interrupting operation of upstream equipment. Thus, a second manifold such as the backup manifold 702 can be arranged to deliver cooling fluid to IT equipment in the IT rack in response to a leak (e.g., a leak detected from the main manifold).

[0050] Figure 8Examples of IT racks and chassis with segmented cooling system architectures according to some embodiments are shown; however, it should be understood that different variations may be implemented. IT rack 900 may contain one or more servers, each with one or more processing units attached to the bottom of any of the aforementioned cooling equipment.

[0051] The rack 900 includes, but is not limited to, a cooling system 940, a rack management unit (RMU) 902 (optional), and one or more server blades 903A-903D (collectively referred to as server blades 903). The cooling system 940 may be any implementation of the cooling system described herein and includes a segmented delivery manifold 941 having sensors and shut-off valves distributed as described.

[0052] As described in other sections, any of the 903A-903E server blades can be assembled into a chassis to house IT equipment (e.g., main PCB, peripherals, PCI-E devices, etc.). The equipment within the chassis is fluidly connected to the IT rack's liquid system during installation.

[0053] Server blades 903 can be inserted into the server slot array from either the front end 904 or the rear end 905 of the IT rack 900. Note that although only five server blades 903A-903E are shown here, more or fewer server blades can be maintained within the IT rack 900. It should also be noted that the specific locations of the cooling system 940, RMU 902, and server blades 903 are shown for illustrative purposes only; other arrangements or configurations are also possible. Note that the rack 900 can be open to the environment or partially contained within a rack container, provided that the cooling fans can generate airflow from the front to the rear.

[0054] Additionally, for each server blade 903, a fan module is associated with the server blade. In this embodiment, fan modules 931A-931E (collectively referred to as fan modules 931) are associated with server blades 903A-903E, respectively. Each fan module 931 includes one or more cooling fans. The fan modules 931 may be mounted on the rear end of the server blade 903 to generate airflow that flows from the front end 904, travels through the air space of the server blade 903, and exists at the rear end 905 of the rack 900.

[0055] The condenser of cooling system 940 can be connected to external liquid supply / return lines 931-932 to form a primary loop. In some embodiments, if the condenser is outside the IT rack, the external fluid supply / return lines can be connected to ports of cooling system 940. Connectors to external liquid supply / return lines 931-932 can be arranged or mounted on the rear end 905 of rack 900. In some embodiments, liquid supply / return lines 931-932 are connected to a set of room manifolds that connect to an external heat removal system or an external cooling loop. The input and output channels of the cooling system can be connected to liquid manifold 925 to form a secondary loop, which may include a supply manifold to supply coolant to server blades 903 and a return manifold to return warmer liquid to cooling system 940.

[0056] Each server blade 903 may include one or more IT components (e.g., a central processing unit or CPU, a graphics processing unit (GPU), memory, and / or storage devices). Each IT component can perform data processing tasks, and the IT component may include software installed in the storage device, loaded into memory, and executed by one or more processors to perform the data processing tasks. At least some of these IT components may be attached to the bottom of any cooling device as described above. The server blade 903 may include a host server (referred to as a master node) coupled to one or more computing servers (also referred to as computing nodes, such as CPU servers and GPU servers).

[0057] A host server (with one or more CPUs) typically connects to a client interface via a network (e.g., the Internet) to receive requests for specific services, such as storage services (e.g., cloud-based storage services like backup and / or recovery) or applications to perform certain operations (e.g., image processing, deep data learning algorithms, or modeling, as part of a Software as a Service (SaaS) platform). In response to the request, the host server assigns the task to one or more performance computing nodes or compute servers (with one or more GPUs) managed by the host server. The performance computing servers perform the actual tasks, which may generate heat during operation.

[0058] IT rack 900 may also include an optional RMU 902, which is configured to provide and manage power supplied to server 903, fan module 931, and cooling system 940. RMU 902 may be coupled to a power supply unit (not shown) to manage the power supply unit's power consumption. The power supply unit may include necessary circuitry (e.g., AC-to-DC or DC-to-DC power converters, backup batteries, transformers, or regulators) to power the remaining components of IT rack 900.

[0059] In one implementation, the RMU 902 includes an optimization module 921 and a rack management controller (RMC) 922. The RMC 922 may include monitors to monitor the operational status of various components within the rack 900, such as compute nodes 903, cooling systems 940, and fan modules 931. Specifically, the monitors receive operational data representing the operating environment of the IT rack 900 from various sensors. For example, the monitors may receive operational data representing the temperature of the processor, coolant, and airflow, which may be captured and collected via various temperature sensors. The monitors may also receive data representing the fan power and pump power generated by the fan module 931 and the liquid pump 912, which may be proportional to the speeds of the fan module 931 and the liquid pump 912, respectively. This operational data is referred to as real-time operational data. Note that the monitors may be implemented as a separate module within the RMU 902.

[0060] Based on the operational data, optimization module 921 performs optimization using a predetermined optimization function or optimization model to derive a set of optimal fan speeds for fan module 931 and optimal pump speeds for liquid pump 912, minimizing the total power consumption of liquid pump 912 and fan module 931, while ensuring that the operational data associated with the cooling fans of liquid pump 912 and fan module 931 are within their respective design specifications. Once the optimal pump speed and optimal fan speed are determined, RMC 922 configures the cooling fans of liquid pump 912 and fan module 931 based on the optimal pump speed and fan speed.

[0061] As an example, based on the optimal pump speed, RMC 922 communicates with the pump controller of cooling system 940 to control the speed of liquid pump 912, which in turn controls the flow rate of coolant supplied to liquid manifold 925 for distribution to at least some server blades 903. Thus, operating conditions and corresponding cooling equipment performance are regulated. Similarly, based on the optimal fan speed, RMC 922 communicates with each fan module 931 to control the speed of each cooling fan in fan module 931, which in turn controls the airflow rate of fan module 931. Note that each fan module 931 can be individually controlled with its specific optimal fan speed, and different fan modules and / or different cooling fans within the same fan module can have different optimal fan speeds.

[0062] Note that some or all of the IT components of server 903 may be attached to any of the aforementioned cooling devices via air cooling using a heat sink or liquid cooling using cold plates. One server may utilize air cooling while another server may utilize liquid cooling. Alternatively, one IT component of a server may utilize air cooling while another IT component of the same server may utilize liquid cooling. Liquid-cooled IT devices may have one or more cold plates attached to the electronic components of the IT device, thereby transferring heat between the IT device and the fluid.

[0063] It should be understood that some features described and shown in the accompanying drawings may vary without departing from the scope of this disclosure. Furthermore, additional valves or auxiliary units may be added to the cooling system for additional features. Furthermore, different types of valves (e.g., three-way valves) may be implemented in the cooling system to achieve the same result. In some embodiments, the controller may adjust the opening of any valve between fully closed (0%) and fully open (100%) as described herein.

[0064] Some implementations may include a non-transitory machine-readable medium (e.g., microelectronic memory) on which instructions are stored, programming one or more data processing units (generally referred to herein as "processors") to perform the operations described herein. In some implementations, some of these operations may be performed by specific hardware components containing hard-wired logic. These operations may alternatively be performed by any combination of programmable data processing units and fixed hard-wired circuit components.

[0065] In the foregoing description, embodiments of the present disclosure have been described with reference to specific exemplary embodiments. However, it will be apparent that various modifications may be made thereto without departing from the broader spirit and scope of the present disclosure as set forth in the appended claims. Therefore, the description and drawings should be considered illustrative rather than restrictive.

[0066] Although certain aspects have been described and illustrated in the accompanying drawings, it should be understood that these aspects are merely illustrative of the broad scope of this disclosure and not limiting, and that this disclosure is not limited to the specific configurations and arrangements shown and described, as various other modifications will be apparent to those skilled in the art. Therefore, this specification should be considered illustrative rather than restrictive.

[0067] In some aspects, this disclosure may include language, such as "at least one of [element A] and [element B]". This language can refer to one or more elements. For example, "at least one of A and B" can refer to "A", "B", or "A and B". Specifically, "at least one of A and B" can refer to "at least one A and at least one B" or "at least one of A or B". In some aspects, this disclosure may include language, such as "[element A], [element B], and / or [element C]". This language can refer to any element or any combination thereof. For example, "A, B, and / or C" can refer to "A", "B", "C", "A and B", "A and C", "B and C", or "A, B, and C".

Claims

1. A liquid cooling system for an information technology (IT) rack, comprising: A liquid manifold comprising multiple sections, wherein each of the multiple sections has one or more shut-off valves; One or more leak detection sensors; A controller having one or more processors is configured to: Receive the status of one or more leak detection sensors; and In response to a leak being detected by a first leak detection sensor among the one or more leak detection sensors, a first shut-off valve among the one or more shut-off valves associated with the first leak detection sensor is controlled to the closed position; It also includes a second manifold, which is arranged to deliver cooling fluid to the IT equipment in the IT rack in response to the leak; The second manifold is arranged in the same way as the liquid manifold, and shuts off the section if a leak is detected. The second manifold seamlessly directs fluid to the section without interrupting operation of upstream equipment.

2. The liquid cooling system according to claim 1, wherein, Each of the multiple sections delivers cooling fluid to the corresponding IT equipment that is vertically stacked together on the IT rack.

3. The liquid cooling system according to claim 2, wherein, In the IT rack, the IT equipment of higher importance is positioned above the IT equipment of lower importance.

4. The liquid cooling system according to claim 3, wherein, The processor is configured to control a second shut-off valve among the one or more shut-off valves to open when a second leak detection sensor among the one or more leak detection sensors does not detect a leak. The second shut-off valve is arranged to control flow to a second IT device positioned above a first IT device among the IT devices, which is shut off from receiving fluid by the first shut-off valve among the one or more shut-off valves in the closed position.

5. The liquid cooling system according to claim 4, wherein, The processor is also configured to communicate with the first IT device and the second IT device in the IT device in response to the leak, so as to migrate one or more operations performed by the first IT device in the IT device to the second IT device in the IT device.

6. The liquid cooling system according to claim 1, wherein, Each of the plurality of segments includes a first connector at a first end and a second connector at a second end, wherein the first connector and the second connector are blind-fit connectors or quick-disconnect connectors.

7. The liquid cooling system according to claim 6, wherein, At least one of the first connector and the second connector includes a corresponding shut-off valve among the one or more shut-off valves.

8. The liquid cooling system according to claim 7, wherein, Both the first connector and the second connector include a corresponding shut-off valve from the one or more shut-off valves.

9. The liquid cooling system according to claim 1, wherein, The one or more leak detection sensors are arranged in each of the plurality of sections and in each corresponding IT device to detect whether any leak exists in any of the plurality of sections or in any of the corresponding IT devices.

10. An information technology (IT) rack comprising a liquid cooling system according to any one of claims 1-9.

Images