Prioritized liquid cooling facility for data center server racks

Abstract

A liquid cooling facility is proposed for cooling a server rack that houses a plurality of server clusters containing a heat processing assembly having a liquid cooling unit. The cooling facility includes a liquid distribution loop that supplies cooling liquid to the server cluster and returns heated liquid from the server cluster, where the heat exchanger is fluidly connected to a supply side of the liquid distribution loop. The server rack includes: a high priority server cluster including processing components having a high heat generation temperature profile and fluidly connected in parallel with the heat exchanger; and a standard priority server cluster including a processing component having a low heat generation temperature profile and positioned further downstream along the liquid distribution loop and configured to receive liquid exiting from the high priority server cluster, the high priority server cluster is arranged to preferentially access the supplied cooling liquid relative to the standard priority server cluster.

Description

[0001] Cross - reference to related applications

[0002] This application claims priority to European Patent Application No. 23307381.6, filed on December 22, 2023, entitled "PRIORITIZED LIQUID COOLING ARRANGEMENTS FOR DATACENTER SERVER RACKS", the entire content of which is incorporated herein by reference. Technical field

[0003] This technology relates to liquid - cooling configurations for data - center server racks. Background art

[0004] Data centers are configured to house multiple server racks that accommodate electronic devices such as computer systems (e.g., server components), storage, etc., in an attempt to process large amounts of data nearly in real - time. During operation, the electronic devices in the server racks generate a large amount of heat that must be dissipated to ensure the continuous and efficient operation of the electronic devices. Many cooling solutions have been implemented to address this heat - generation problem, including liquid - cooling (commonly referred to as liquid - block units or water - block units) of heat - generating components by directly mounting liquid - cooling blocks onto certain heat - generating components.

[0005] While water - block units can effectively cool heat - generating components, the implementation of water - block units in server racks typically requires liquid - distribution infrastructure to serve multiple server racks and the large number of electronic devices supported within the server racks. Additionally, some server racks house servers with high - performance electronic - processing components that generate a higher temperature profile, while other servers contain electronic components that do not generate a higher temperature profile.

[0006] It will be appreciated that liquid - cooling distribution infrastructure generally requires the use of a large number of transmission - pipe implementations across the data center, as well as multiple pumps to maintain the necessary coolant flow rate to supply a large number of water - blocks, particularly those for high - performance electronic - processing components. In terms of initial investment and operating costs, the use of a large number of transmission - pipe implementations and multiple pumps is prohibitively expensive for data centers.

[0007] Accordingly, it appears that there is a need to provide a liquid cooling facility for a data center server rack that can mitigate at least some of the excessive cost problems associated with conventional conduit implementations and pumps by providing a configuration that effectively addresses the cooling requirements of both servers with processing components having a high temperature profile due to electronic components and servers with processing components having a standard temperature profile.

[0008] It should be noted that the subject matter discussed in the background section should not be regarded as prior art merely because it is mentioned in the background section. Similarly, the problems mentioned in the background section should not be construed as having been recognized in the prior art. Summary of the Invention

[0009] The object of the present technology is to mitigate at least some of the excessive cost problems prevalent in the prior art.

[0010] According to one aspect of the present technology, there is provided a liquid cooling facility for cooling a server rack that houses a plurality of server clusters, each of the plurality of server clusters including a heat-generating processing component having a liquid cooling unit. The liquid cooling facility includes: a liquid distribution loop configured to supply cooling liquid from a liquid cooling device to the server clusters and return the heated liquid from the server clusters to the cooling device; and one or more heat exchangers fluidly connected to the supply side of the liquid distribution loop. The server rack includes: a high-priority server cluster that includes a processing component having a high heat generation temperature profile and is fluidly connected to the one or more heat exchangers in parallel; and a standard-priority server cluster that includes a processing component having a heat generation temperature profile lower than that of the high-priority server cluster and is positioned downstream of the high-priority server cluster along the liquid distribution loop; and the standard-priority server cluster is configured to receive the liquid exiting the high-priority server cluster. The high-priority server cluster is arranged to preferentially access the supplied cooling liquid relative to the standard-priority server cluster.

[0011] In some aspects, the one or more heat exchangers include a plurality of air-liquid heat exchangers arranged in a parallel configuration relative to the liquid distribution loop.

[0012] In other aspects, the high-priority cluster includes a plurality of clusters arranged in a parallel configuration relative to the liquid distribution loop, and the standard-priority cluster includes a plurality of clusters arranged in a parallel configuration relative to the liquid distribution loop.

[0013] In another aspect, the liquid entering the high-priority cluster is at a first temperature, the liquid leaving the high-priority cluster is at a second temperature higher than the first temperature, the liquid entering the standard-priority cluster is at the second temperature, and the liquid leaving the standard-priority cluster is at a third temperature higher than the second temperature.

[0014] In some aspects, the standard-priority cluster receives the liquid leaving the high-priority cluster and the heat exchanger via a liquid distribution loop. In other aspects, the standard-priority cluster directly receives the liquid leaving the high-priority cluster from a series fluid connection independent of the one or more heat exchangers.

[0015] In yet another aspect, one of the heat exchangers includes a primary side and a secondary side, wherein the primary side and the secondary side of the heat exchanger are fluidly connected to each other in parallel, and the high-priority cluster is fluidly connected in series to either the primary side or the secondary side of a given heat exchanger in the heat exchanger, while remaining in parallel with the remaining heat exchangers.

[0016] Each implementation of the present technology has at least one of the above objects and / or aspects, but not necessarily all of the above objects and / or aspects. It should be understood that some aspects resulting from the attempt to achieve the above objects of the present technology may not satisfy that object and / or may satisfy other objects not specifically listed herein.

[0017] Additional and / or alternative features, aspects, and advantages of the implementations of the present technology will become apparent from the following description, drawings, and appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] For a better understanding of the present technology and other aspects and additional features thereof, reference is made to the following description taken in conjunction with the drawings, in which:

[0019] Figure 1 A functional block diagram of an internal server cluster configuration of a data center server rack according to a non-limiting embodiment of the present disclosure is depicted;

[0020] Figure 2 A functional block diagram of a liquid cooling facility according to some non-limiting embodiments of the present disclosure is depicted, which optimizes cooling efficiency by prioritizing server clusters within a data center rack;

[0021] Figure 3 A functional block diagram of an alternative liquid cooling facility according to some non-limiting embodiments of the present disclosure is depicted, which optimizes cooling efficiency by prioritizing server clusters within a data center rack; and

[0022] Figure 4 Depicts a functional block diagram of another alternative liquid cooling facility in accordance with some non - limiting embodiments of the present disclosure, which optimizes cooling efficiency by prioritizing server clusters within a data center rack. Detailed implementation

[0023] The present disclosure aims to address at least some of the problems associated with conventionally using various transmission pipeline configurations and multiple pumps to supply a liquid flow to water blocks to meet the cooling requirements of a large number of heat - generating components. In particular, the present disclosure presents embodiments of a liquid cooling facility that prioritizes higher - heat - generating servers to optimize the cooling of such servers while streamlining the liquid cooling transmission pipeline facility.

[0024] The examples and conditional language recited herein are principally intended to assist the reader in understanding the principles of the technology and are not intended to limit the scope of the technology to these specifically recited examples and conditions. It will be understood that those skilled in the art can devise various arrangements that, although not explicitly described or shown herein, still embody the principles of the technology.

[0025] In addition, for the sake of helping understanding, the following description may describe relatively simplified implementations of the technology. As those skilled in the art will understand, the various implementations of the technology may have greater complexity.

[0026] In some cases, examples of modifications that are considered helpful to the technology may also be set forth. This is done solely to assist understanding and is likewise not intended to limit the scope of the technology or to delineate the boundaries of the technology. These modifications are not an exhaustive list, and those skilled in the art can make other modifications while still remaining within the scope of the technology. Moreover, in the absence of setting forth examples of modifications, it should not be construed that modifications are not possible and / or that the described content is the only way to implement the element of the technology.

[0027] Furthermore, all statements herein reciting principles, aspects, and implementations of the technology, as well as specific examples of the technology, are intended to cover structural and functional equivalents of the technology, whether currently known or developed in the future. Thus, for example, those skilled in the art will understand that any block diagrams herein represent a conceptual view of an illustrative system embodying the principles of the technology.

[0028] Based on the above, we will now consider some non - limiting examples to illustrate the implementations of the various inventive aspects of the present disclosure.

[0029] Figure 1 FIG. Figure 1 illustrates a functional block diagram of a representative internal server cluster 10 of a data center server rack according to a non-limiting embodiment of the present disclosure. The schematic configuration of the server cluster 10 includes a plurality of server sub-clusters 20 to 2M, where each server sub-cluster 20 to 2M correspondingly includes a plurality of data processing components 20A to 20N to 2MA to 2MN containing heat-generating electronic processing components.

[0030] As shown, each of the data processing components 20A to 20N to 2MA to 2MN is coupled with at least one corresponding liquid cooling block unit 20A1 to 20N1 to 2MA1 to 2MN1, which is arranged to be in direct thermal contact with the heat-generating electronic processing components. Each of the liquid cooling block units 20A1 to 20N1 to 2MA1 to 2MN1 is configured to have internal pipes (not shown) to accommodate a circulating flow of cooling liquid channelized through the internal pipes. The supplied cooling liquid is provided by a cooling liquid supply device, and the supplied cooling liquid is serially conveyed via an internal server cluster liquid circulation channel 30 to each of the liquid cooling block units 20A1 to 20N1 to 2MA1 to 2MN1 to absorb heat energy from the heat-generating electronic components and discharge the heated liquid from the internal server cluster liquid circulation channel 30.

[0031] Considering the above configuration of the internal data center server cluster 10, Figure 2 FIG. Figure 2 illustrates a functional block diagram of a liquid cooling facility 100 according to a non-limiting embodiment of the present disclosure, which optimizes the overall cooling efficiency by prioritizing the server clusters of a data center rack. Although the embodiments and implementations described herein are directed to a data center rack, it will be understood that extending such concepts to cover multiple data center racks is clearly within the scope of the present technology.

[0032] As shown, the liquid cooling facility 100 includes a single liquid distribution loop 105, which is configured to have: a supply side for supplying cooling liquid from a liquid cooling device 170 to server clusters 120 to 12N, 130 to 13M, 140 to 14P, 150 to 15L of the rack; and a return side for returning the heated liquid from the server clusters 120 to 15L to the cooling device 170 for re-cooling and re-circulating back to the server clusters 120 to 15L. As noted above with respect to Figure 1 each of the server clusters 120 to 15L includes data processing components (see, for example, Figure 1from 20A to 20N to 2MA to 2MN), the data processing component is combined with at least one corresponding liquid cooling block unit for direct thermal contact with the heat - generating electronic components of the data processing component to dissipate heat from the heat - generating electronic components.

[0033] The liquid distribution loop 105 is configured to have a liquid distribution inlet 101 along the supply side for supplying cooling liquid to the fluid - coupled server clusters 120 to 15L, and the liquid distribution loop 105 is configured to have a liquid distribution outlet 102 along the return side for receiving the heated liquid from the server clusters 120 to 15L and returning the heated liquid to the cooling device 170 for recooling and recycling back to the server clusters 120 to 15L. The liquid distribution loop 105 can be constructed of a flexible material (e.g., rubber, plastic, etc.), a rigid material (e.g., metal, PVC piping, etc.), or any combination of the foregoing. It will be understood that the liquid being conveyed can include water, alcohol, or any suitable liquid capable of maintaining a sufficient cooling temperature.

[0034] The cooling device 170 can include a dry cooler unit 172 configured to process and recondition the liquid received from the server rack to provide cooling liquid for recycling back to the server clusters 120 to 15L via the liquid distribution loop 105. The cooling device 170 can also include a pump 174 configured to provide the pressure increase and volumetric flow rate required for the cooling liquid from the dry cooler unit 172 throughout the liquid distribution loop 105.

[0035] The liquid cooling facility 100 also includes a plurality of air - liquid heat exchangers (ALHEX) 110 to 114. In the illustrated embodiment, the ALHEX 110 to 114 are connected in parallel via the liquid cooling loop 105, and the ALHEX 110 to 114 are also fluid - coupled to the server clusters 120 to 15L via the liquid cooling loop 105. However, it will be understood that without departing from the concepts of the disclosed technology, the ALHEX 110 to 114 can be fluid - interconnected in other configurations, for example, the ALHEX 110 to 114 can be fluid - interconnected in series via the liquid cooling loop 105.

[0036] ALHEX 110 to 114 are used to sufficiently cool the ambient air around server clusters 120 to 15L. ALHEX 110 to 114 can achieve any suitable configuration for reducing the liquid temperature through a supplied air flow (e.g., through a compact fan), such as internal cooling coils, heat extraction air flow fins, etc. For example, ALHEX 110 to 114 can be disposed on the rear door of the rack carrying server clusters 120 to 15L to directly cool the air that leaves server clusters 120 to 15L and is warmed by air-cooled components in the server clusters 120 to 15L.

[0037] As described above, certain servers containing high-performance electronic processing components experience a higher heat generation temperature profile compared to other servers. Additionally, generally, liquid-cooled components arranged in parallel reduce the fluid pressure drop experienced by the liquid distribution loop 105, thereby allowing the pump 174 to operate efficiently to provide a flow rate sufficient to circulate throughout the liquid distribution loop 105.

[0038] Taking these operating factors into account, the liquid cooling facility 100 provides the following configuration: in this configuration, one or more server clusters incorporating high-performance electronic processing components are designated as "high-priority" server clusters, while server clusters not incorporating high-performance electronic processing components are designated as "standard-priority" server clusters. In some implementations, the electronic processing components of the servers are enclosed inside a housing that is thermally connected to a water block; the high-priority server clusters may require an operational upper casing temperature below approximately 60 °C, while the standard-priority server clusters may require an operational upper casing temperature in the range of approximately 70 °C to 85 °C. Additionally, the high-priority server clusters are strategically positioned in parallel with the ALHEX to optimize the cooling of the high-priority server clusters while effectively maintaining a satisfactory flow rate circulating throughout the liquid distribution loop 105.

[0039] That is, as Figure 2As shown, the liquid cooling facility 100 configures one or more high - priority server clusters 120 through 12N and 160 through 16K to be arranged in parallel with the ALHEXs 110 through 114 via the liquid distribution loop 105 and fluidly connected to the ALHEXs 110 through 114. This configuration also positions one or more high - priority server clusters 120 through 12N and 160 through 16K close to the ALHEXs 110 through 114. In this configuration, one or more high - priority server clusters 120 through 16K are positioned to receive first the "coldest" cooling liquid provided by the liquid distribution loop 105 to specifically address the particular cooling requirements of the high - priority server clusters 120 through 12N to 160 through 16K.

[0040] After the "coldest" cooling liquid has circulated internally through each of the data - processing components of the one or more high - priority server clusters 120 through 16K, the liquid exiting the high - priority server clusters 120 through 16K is then supplied via the liquid distribution loop 105 to the "standard - priority" server clusters 130 through 13M, 140 through 14P, 150 through 15L for cooling within these "standard - priority" server clusters. Similar to the high - priority server clusters 120 through 16K, the standard - priority server clusters 130 through 15L are arranged in parallel via the liquid cooling loop 105.

[0041] Although the liquid exiting the high - priority server clusters 120 through 16K via the liquid distribution loop 105 will typically be warmer compared to the "coldest" cooling liquid due to the corresponding heat - generating processing components, the temperature of the "warm" liquid will generally remain at a level sufficient to cool the standard - priority server clusters 130 through 15L. This is due to the fact that the standard - priority server clusters 130 through 15L do not employ high - performance electronic processing components that have a higher heat - generation temperature profile and require a lower temperature (i.e., "coldest") cooling liquid. In some implementations, the coldest liquid temperature can be in the range of about 25°C to 35°C, while the warm liquid temperature can be in the range of about 35°C to 45°C.

[0042] After the "warm" liquid has circulated internally through each of the data - processing components and cooling units of the standard - priority server clusters 130 through 15L, the liquid exiting the standard - priority server clusters 130 through 15L is heated. The "heated" liquid is conveyed to the return side of the liquid distribution loop 105 to return to the cooling device 170, thereby re - cooling the "heated" liquid and recycling it back to all of the server clusters 120 through 15L. In some implementations, the heated liquid temperature can be in the range of about 45°C to 65°C.

[0043] Figure 3 FIG. 2 depicts a functional block diagram of an alternative liquid cooling facility 200 in accordance with some non - limiting embodiments of the present disclosure, which also optimizes cooling efficiency by prioritizing server clusters within a data center rack. Since the liquid cooling facility 200 includes components with similar reference numerals to those of the facility 100, for the sake of brevity, the detailed description of these components will not be repeated unless necessary for understanding the embodiment.

[0044] Similar to the liquid cooling facility 100, the liquid cooling facility 200 includes high - priority server clusters 120 through 12N, a plurality of standard - priority server clusters 130 through 15L, and a liquid distribution loop 105 that conveys a circulating flow of cooling liquid through corresponding processing components and cooling units of the server clusters.

[0045] The liquid cooling facility 200 also includes a plurality of ALHEXs 110 through 114, which are configured in parallel with respect to the liquid distribution loop 105. As described above, the ALHEXs 110 through 114 are used to sufficiently cool the ambient air around the server clusters 120 through 15L (e.g., by being disposed on the rear door to directly cool the air that has been warmed by air - cooled components within the server clusters 120 through 15L as it exits the server clusters 120 through 15L).

[0046] However, different from the liquid cooling facility 100, the liquid cooling facility 200 divides the standard - priority server clusters 130 through 15L into two or more standard - priority server groups. For example, the liquid cooling facility 200 divides the standard - priority server clusters 130 through 15L into a first group and a second group. The first group is clusters 130 through 13M and 150 through 15L; the second group is clusters 140 through 14P. As shown, the first standard - priority server group arranges clusters 130 through 13M and 150 through 15L in parallel with each other, where the clusters are fluid - coupled to the liquid distribution loop 105 to receive the warmed cooling liquid provided by the ALHEXs 110 through 114.

[0047] Accordingly, the second standard priority server groups (i.e., clusters 140 to 14P) receive in series and directly the liquid exiting from the high priority server clusters 120 to 12N for cooling purposes. That is, the high priority server clusters 120 to 12N are fluidly coupled on the supply side to the liquid distribution loop 105 to receive the "coldest" cooling liquid and circulate the fluid internally through the respective processing components and cooling units of the entire high priority server clusters 120 to 12N. Additionally, as described above, although the liquid exiting from the high priority server clusters 120 to 12N will typically be warmer than the "coldest" cooling liquid due to the respective high heat generating processing components, the temperature of the "warm" liquid will typically still be at a level sufficient to cool the second standard priority server groups (i.e., clusters 140 to 14P). For some cluster configurations, the liquid cooling facility 200 provides advantages that are faster and / or easier to implement compared to the cooling facility 100.

[0048] Figure 4 FIG. depicts a functional block diagram of another alternative liquid cooling facility 300 in accordance with some non-limiting embodiments of the present disclosure, which also optimizes cooling efficiency by prioritizing server clusters within a data center rack. Since the liquid cooling facility 300 includes components with similar reference numerals to those of the facility 100, the detailed description of these components will not be repeated for the sake of brevity, unless necessary for understanding the embodiment.

[0049] Similar to the liquid cooling facility 100, the liquid cooling facility 300 includes: high priority server clusters 120 to 12N, a plurality of standard priority server clusters 130 to 15L; a liquid distribution loop 105 for conveying a circulating flow of cooling liquid throughout the server clusters; and a plurality of ALHEXs 110 to 114 arranged in parallel with respect to the liquid distribution loop 105.

[0050] As described above, the ALHEXs 110 to 114 are used to sufficiently cool the ambient air around all the server clusters 120 to 15L (e.g., by being disposed on the back door to directly cool the air exiting the server clusters 120 to 15L that has been warmed by the air-cooled components within these server clusters 120 to 15L). It will be understood that each of the ALHEXs 110 to 114 may include a primary cooling side and a secondary cooling side, where each side may be operated independently and have complementary cooling effects.

[0051] As described above, unlike the liquid cooling facility 100, the liquid cooling facility 300 utilizes the following ability: ALHEX 114 can be divided into an ALHEX primary side 114A and a secondary side 114B, wherein the high-priority server clusters 120 to 12N are fluidly coupled to one side of the ALHEX 114B in series.

[0052] Specifically, as shown, the high-priority server clusters 120 to 12N receive the "coldest" cooling liquid via the liquid distribution loop 105. Then, the cooling liquid circulates internally through the respective processing components and cooling units of the entire high-priority clusters 120 to 12N, resulting in "warm" liquid leaving the high-priority server clusters 120 to 12N. The warm liquid is serially conveyed to the ALHEX side 114B for cooling, and the cooled liquid is then supplied to the liquid distribution loop 105 to liquid-cool the standard-priority server clusters 130 to 15L. With this configuration, the liquid cooling facility 300 provides the advantage of maximizing the liquid flow rate through the high-priority clusters.

[0053] Air flows along Figure 4 the arrow direction depicted in. When entering the primary side 114A of the ALHEX, the temperature of the air flow is relatively high due to the heat generated in the server racks. Thus, the temperature of the liquid flowing through the primary side 114A of the ALHEX may be further increased, thus being continuously exposed to the air flow. The air flow between the primary side 114A and the secondary side 114B of the ALHEX may have a stratified and non-uniform temperature profile. The same air flow immediately reaches the secondary side 114B of the ALHEX. In this implementation form, the cold liquid is directly received by the secondary side 114B of the ALHEX from the inlet 101. Therefore, the air flow leaves the secondary side 114B of the ALHEX at a reduced temperature and can thus be redirected to the server racks to cool the server racks.

[0054] In this way, each of the presented liquid cooling facilities 100, 200, 300 provides a configuration that optimizes the overall cooling efficiency by separately addressing the liquid cooling requirements of the high-priority server clusters and the standard-priority server clusters.

[0055] Modifications and improvements to the above-described implementations of the technology may become apparent to those skilled in the art. The foregoing description is intended to be exemplary and not restrictive. Therefore, the scope of the technology is intended to be limited only by the scope of the appended claims.

Claims

1. A liquid cooling facility (100, 200, 300) for cooling a server rack, wherein the server rack accommodates a plurality of server clusters (120-12N, 160-16K, 130-13M, 140-14P, 150-15L), each of the server clusters comprising a heat treatment component having a liquid cooling unit, the liquid cooling facility (100, 200, 300) comprising: a liquid distribution circuit (105) configured to supply cooling liquid from a liquid cooling device (170) to the server cluster (120-15L) and to return heated liquid from the server cluster (120-15L) to the liquid cooling device (170); one or more heat exchangers (110-114), the one or more heat exchangers (110-114) being fluidly connected to a supply side of the fluid distribution circuit (105); The server rack comprises: a high priority server cluster (120-12M, 160-16K), the high priority server cluster (120-12M, 160-16K) containing processing components having a high exothermic temperature profile, and the high priority server cluster (120-12M, 160-16K) is fluidly connected in parallel to the one or more heat exchangers, and a standard priority server cluster (130-15L), the standard priority server cluster (130-15L) comprising processing components having a heating temperature curve lower than the heating temperature curve of the high priority server cluster, and the standard priority server cluster (130-15L) being positioned further downstream along the liquid distribution loop than the high priority server cluster; and the standard priority server cluster (130-15L) being configured to receive liquid leaving the high priority server cluster, Wherein, the high priority server cluster is arranged to have priority access to the supplied cooling liquid relative to the standard priority server cluster.

2. The liquid cooling facility according to claim 1, wherein: The one or more heat exchangers include an air-to-liquid heat exchanger.

3. The liquid cooling facility according to claim 1 or 2, wherein: The one or more heat exchanges include a plurality of heat exchangers arranged in a parallel configuration relative to the liquid distribution circuit (105).

4. The liquid cooling facility according to any one of claims 1 to 3, further comprising a pump (174) configured to control the flow of liquid circulating throughout the liquid distribution circuit (105).

5. The liquid cooling facility according to any one of claims 1 to 4, wherein: The liquid cooling device (170) also includes a dry cooler unit (172).

6. The liquid cooling facility according to any one of claims 1 to 5, wherein: The standard priority server cluster is arranged in a parallel configuration relative to the liquid distribution circuit (105).

7. The liquid cooling facility according to any one of claims 1 to 6, wherein: Liquid entering the high priority server cluster is at a first temperature, and liquid exiting the high priority server cluster is at a second temperature, the second temperature being higher than the first temperature.

8. The liquid cooling facility according to any one of claims 1 to 7, wherein: Liquid entering the standard priority server cluster is at the second temperature, and liquid exiting the standard priority server cluster is at a third temperature, the third temperature being higher than the second temperature.

9. The liquid cooling installation according to any one of claims 1 to 8, wherein: The standard priority server cluster receives liquid exiting from the high priority server cluster and the one or more heat exchangers via the liquid distribution loop.

10. The liquid cooling installation according to any one of claims 1 to 9, wherein: The standard priority server cluster receives liquid exiting from the high priority server cluster directly from a serial fluid connection independent of the one or more heat exchangers (110-114).

11. The liquid cooling installation according to any one of claims 1 to 10, wherein: A given heat exchanger of the one or more heat exchangers includes a primary side and a secondary side.

12. The liquid cooling facility according to claim 11, wherein: The primary side and the secondary side of a given one of the one or more heat exchangers are fluidly connected to each other in parallel.

13. The liquid cooling facility according to claim 12, wherein: The high priority server cluster is fluidly connected in series with the primary side or the secondary side of a given one of the one or more heat exchangers.

14. The liquid cooling facility according to claim 13, wherein: The high priority server cluster and the serially connected heat exchangers are arranged in parallel with the remaining heat exchangers.

Images