The present application claims priority to EP Application No 23305473.3, entitled “SERIALIZED LIQUID COOLING ARRANGEMENTS FOR DATACENTER SERVER RACKS”, filed Mar. 31, 2023, the entirety of which is incorporated herein by reference.
The present technology relates to datacenter server rack configurations.
Datacenters are configured to house multitudes of server racks containing electronic equipment, such as computer systems (e.g., server assemblies), memory banks, etc. in efforts to process vast amounts of data in near real time. During operations, the electronic equipment of the server racks generates a significant amount of heat that must be dissipated in order to ensure continued efficient operation of the electronic equipment. Many cooling solutions have been implemented to address this heating issue, including the liquid cooling of heat-generating components by way of liquid cooling blocks directly mounted onto certain heat-generating components (often referred to as liquid or water block units).
Although water block units are capable of efficiently cooling the heat-generating components, their implementation in server racks typically requires a liquid distribution infrastructure to service the multitude of server racks and the vast number of electronic equipment supported therein. Such liquid distribution infrastructures conventionally require the use of relatively large and/or heavy piping conduit configurations and large capacity pumps to maintain the necessary liquid flow rates that supply the water blocks to service the cooling needs of the vast number of corresponding heat-generating components. It will be appreciated that the use of such piping conduit configurations and large pumps can be prohibitively costly for datacenters, in terms of initial investments and operating costs. Such piping conduit configurations and large pumps inherently occupy large footprints which may reduce a productivity (e.g. server per unit area of datacenter floor surface).
As a result, it appears to be desirable to provide a liquid cooling arrangement for datacenter server racks that can alleviate at least some of the cost prohibitive issues regarding conventional piping conduit configurations.
It is to be noted that the subject matter discussed in the background section should not be assumed to be prior art merely as a result of its mention in the background section. Similarly, the issues mentioned in the background section should not be interpreted as having been recognized in the prior art.
It is an object of the present technology to alleviate at least some of the cost prohibitive issues that prevail in the prior art.
According to one aspect of the present technology, there is provided a liquid cooling arrangement for cooling heat-generating components of a datacenter server rack that includes a liquid cooling loop configured to convey a cooling liquid, a plurality of server clusters each server cluster having a plurality of server assemblies that incorporate at least one respective liquid cooling unit, and a plurality of heat exchangers fluidly connected to the liquid cooling units of the plurality of server clusters via the liquid cooling loop. The heat exchangers configured to receive the heated cooling liquid discharged from the respective liquid cooling units and cool the received heated cooling liquid along with a pump fluidly coupled to the plurality of heat exchangers via the liquid cooling loop, wherein at least a portion of the liquid cooling loop comprises a serialized configuration that serially couples the server clusters and/or serially couples the server clusters and the heat exchangers.
In some embodiments, the liquid cooling units of the server assemblies of the same server cluster are fluidly connected in parallel to one another.
In some embodiments, the server clusters are fluidly connected in series with one another and the heat exchangers are fluidly connected in parallel with one another.
In some embodiments, each of the server clusters is fluidly connected in series to a corresponding heat exchanger.
In some embodiments, the server clusters are fluidly connected in parallel with one another.
In some embodiments, the liquid cooling units of the server assemblies of a same server cluster are fluidly connected in series to one another.
In some embodiments, each of the server clusters is fluidly connected in series to a corresponding heat exchanger.
In some embodiments, the server clusters are fluidly connected in series with one another and the heat exchangers are fluidly connected in series with one another.
In some embodiments, the liquid cooling units of the server assemblies of a same server cluster are fluidly connected in series to one another.
In some embodiments, the server clusters are fluidly connected in series with one another and the heat exchangers are fluidly connected in series with one another.
According to another aspect of the present technology, there is provided a a liquid cooling arrangement for cooling heat-generating components of a datacenter server rack that comprises a liquid cooling loop configured to convey a cooling liquid, a plurality of server clusters, each server cluster including a plurality of server assemblies that incorporate at least one respective liquid cooling unit configured to collect at least a portion of a thermal energy generated by a heat-generating component of the server assembly, and a pump fluidly coupled to the liquid cooling units of the plurality of server clusters via the liquid cooling loop, the pump configured to convey the cooling liquid in the liquid cooling loop, wherein at least a portion of the liquid cooling loop comprises a serialized configuration that serially couples the server clusters.
In accordance with this other aspect, in some embodiments, the liquid cooling arrangement further comprises a plurality of heat exchangers fluidly connected to the liquid cooling units of the plurality of server clusters via the liquid cooling loop, the heat exchangers configured to receive the heated cooling liquid discharged from the respective liquid cooling units of the server assemblies and cool the received heated cooling liquid.
Additionally, in some embodiments, at least a portion of the liquid cooling loop comprises a serialized configuration that serially couples the server clusters and heat exchangers.
Implementations of the present technology each have at least one of the above-mentioned objects and/or aspects, but may not necessarily have all of them. It should be understood that some aspects of the present technology that have resulted from attempting to attain the above-mentioned object may not satisfy this object and/or may satisfy other objects not specifically recited herein.
Additional and/or alternative features, aspects and advantages of implementations of the present technology will become apparent from the following description, the accompanying drawings and the appended claims.
For a better understanding of the present technology, as well as other aspects and further features thereof, reference is made to the following description which is to be used in conjunction with the accompanying drawings, where:
The instant disclosure is directed to addressing at least some of the issues associated with conventional large/heavy piping conduit configurations that supply the liquid flows to the water blocks to adequately service the cooling needs of the vast number of corresponding heat-generating components. In particular, the instant disclosure presents various embodiments of serialized liquid cooling arrangements for server racks that facilitate the use of smaller, lighter, and streamlined piping conduit configurations.
The examples and conditional language recited herein are principally intended to aid the reader in understanding the principles of the present technology and not to limit its scope to such specifically recited examples and conditions. It will be appreciated that those skilled in the art may devise various arrangements that, although not explicitly described or shown herein, nonetheless embody the principles of the present technology.
Furthermore, as an aid to understanding, the following description may describe relatively simplified implementations of the present technology. As persons skilled in the art would understand, various implementations of the present technology may be of a greater complexity.
In some cases, what are believed to be helpful examples of modifications to the present technology may also be set forth. This is done merely as an aid to understanding, and, again, not to define the scope or set forth the bounds of the present technology. These modifications are not an exhaustive list, and a person skilled in the art may make other modifications while nonetheless remaining within the scope of the present technology. Further, where no examples of modifications have been set forth, it should not be interpreted that no modifications are possible and/or that what is described is the sole manner of implementing that element of the present technology.
Moreover, all statements herein reciting principles, aspects, and implementations of the present technology, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof, whether they are currently known or developed in the future. Thus, for example, it will be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present technology.
With these fundamentals in place, we will now consider some non-limiting examples to illustrate the implementations of the various inventive aspects of the present disclosure.
In particular,
Each of the server clusters 110, 112 . . . 11M includes a plurality of server assemblies 110A-110N, 112A-112N . . . 11MA-11MN that are arranged in a parallel manner. As noted above, the server assemblies 110A-110N, 112A-112N . . . 11MA-11MN contain heat generating electronic components.
Accordingly, each of the parallel server assemblies 110A-110N, 112A-112N . . . 11MA-11MN incorporates at least one respective liquid cooling unit 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1, correspondingly arranged in parallel, for the direct thermal contact liquid cooling of the heat generating electronic components. That is, each of the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 is configured as liquid-cooled heat sink conduit block that is thermally coupled, either directly or indirectly, to the heat-generating electronic components, such that cooling liquid is circulated through internal liquid conduits of the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 to absorb the heat from the heat-generating electronic components and discharge the heated liquid therefrom.
For example, each one of the server clusters 110, 112 . . . 11M may include a first manifold that, in use, receives the cooling liquid and feeds the cooling liquid to the plurality of liquid cooling units of the server cluster in parallel. A second manifold may be provided downstream said plurality of liquid cooling units to receive the cooling liquid from the plurality of liquid cooling units.
The liquid cooling arrangement 100 further includes a plurality of air-to-liquid heat exchangers (ALHEXs) 120, 122 . . . 12M for cooling the cooling liquid. More specifically, in an implementation, each of the air-to-liquid heat exchangers (ALHEXs) 120, 122 . . . 12M defines an exchanger internal fluid conduit that forms a part of the cooling loop 150. Therefore, each of the air-to-liquid heat exchangers (ALHEXs) 120, 122 . . . 12M has an inlet through which, in use, the cooling liquid flows into the exchanger internal fluid conduit, and an outlet through which, in use, the cooling liquid is discharged from the exchanger internal fluid.
The ALHEXs 120, 122 . . . 12M of the liquid cooling arrangement 100 are fluidly connected in parallel with one another. Namely, the internal fluid conduits of the air-to-liquid heat exchangers (ALHEXs) 120, 122 . . . 12M of the liquid cooling arrangement 100 are fluidly connected in parallel. The ALHEXs 120, 122 . . . 12M of the liquid cooling arrangement 100 are also fluidly coupled to server clusters 110, 112 . . . 11M via the liquid cooling loop 150. The ALHEXs 120, 122 . . . 12M function to sufficiently air cool the heated liquid received by the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 for redirection back to the server clusters 110, 112 . . . 11M. The ALHEXs 120, 122 . . . 12M may embody any suitable configuration that reduces liquid temperatures through supplied air flow, such as, internal cooling coils, heat extracting air flow fins, etc. The ALHEXs 120, 122 . . . 12M may be, for example and without limitations, disposed on rear doors of a rack hosting the server clusters 110, 112 . . . 11M.
The liquid cooling arrangement 100 additionally includes at least one pump 130 that is fluidly connected to the server rack liquid cooling loop 150. The pump 130 is configured to receive the cooling liquid from the ALHEXs 120, 122 . . . 12M, via the server rack liquid cooling loop 150, and functions to forcibly provide the necessary circulatory flow rate of the cooling liquid through the server rack liquid cooling loop 150, in order to service the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of server clusters 110, 112 . . . 11M.
It is to be noted that the thermal power exchanged inside each cooling element of the server rack liquid cooling loop 150 capable of recovering or rejecting heat, including the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1, and the ALHEXs 120, 122 . . . 12M, as well as a group containing such cooling elements, including the server clusters 110, 112 . . . 11M, may be determined by the following relationship:
Q=m·C·ΔT, in which: (Eq. 1)
It is to be noted that the relationship of Eq. 1 applied to the ALHEXs 120, 122 . . . 12M is valid for both air and liquid, and while neglecting any losses (e.g., by heat radiation), the thermal power exchanged by the liquid is equaled to the opposite number of the thermal power exchanged by the air. Thus, it exists a relationship between the liquid temperatures difference ΔTliquid and the air temperatures difference ΔTair.
Another parameter of interest are the “pinch” values of the ALHEXs 120, 122 . . . 12M. That is, each of the ALHEXs 120, 122 . . . 12M operate in a counter current configuration and thus has a “hot side” and a “cold side”. So, for the hot side, the pinch value ΔThotpinch is defined as the difference between the temperature of the hot liquid entering the ALHEX to be cooled by and the temperature of the hot air exiting the ALHEX. And, for the cold side, the pinch value ΔTcoldpinch is defined as the difference between the temperature of the cooled liquid exiting the ALHEX and the temperature of cooled fresh air entering the ALHEX. Both of the pinch values ΔThotpinch, ΔTcoldpinch are positive numbers. A low pinch value ΔTcoldpinch can benefit to the components to be cooled by the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1. During operations, depending on the thermal power exchanged and the liquid and air mass flow rates, pinch value ΔThotpinch may act as a “limiting pinch” that affects or conditions negatively a lower value of the ΔTcoldpinch.
As indicated above, the liquid cooling arrangement 100 employs a configuration that combines serialized and parallelized liquid cooled components to achieve efficient circulatory flow rates and maintain adequate liquid cooling temperatures. As a general rule, liquid cooled components arranged in series, increase the fluid pressure drop in the liquid cooling loop 150 thereby inducing the pump 130 to operate at a lower flow rate. In contrast, liquid cooled components arranged in parallel, decrease the fluid pressure drop in the liquid cooling loop 150 thereby allowing the pump 130 to operate at a higher flow rate. Regarding the usual hydraulic characteristic curve of the pump 130, arranging the liquid cooled components in series will more likely increase the flow rate inside each of them, in comparison to the local flow rate reachable in a similar loop where the liquid cooled components are arranged in parallel, and the global flow rate is distributed between each parallelized branch. In most similar applications, the preference is generally given to parallelized configuration of identical components, with, for example, parallelized server clusters 110, 112 . . . 11M, parallelized liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1, parallelized ALHEXs 120, 122 . . . 12M, for different reasons, including the ease of maintenance and the uniformization of the operating points and thermal behavior for the identical components.
It is to be noted that as the liquid flow is circulated the through liquid cooling loop 150, the ALHEXs 120, 122 . . . 12M may lose efficiency if their local liquid flow rate is reduced to a lower value. In an effort to achieve a balance between maintaining ALHEX efficiency and providing suitable circulatory flow rates, reference is made to the ΔT indicating the fluid temperature differences between the input and output sides of the serialized server clusters (see, Eq. 1), from the input of the first server cluster 110 to the output of the last server cluster 11M, as well as the pinch values ΔThotpinch, ΔTcoldpinch, and the temperature differences on the ALHEXs ΔTliquid, ΔTair, noted above.
In particular, as the circulatory liquid flow rate on the liquid input side of the ALHEX is lowered, ΔTliquid between the liquid input and output of the ALHEX increases until it becomes closer to the ΔTair between the air output and input on the air side of the ALHEX. In so doing, ΔThotpinch is no longer limiting and the value of ΔTcoldpinch is reduced along with the minimum temperature of the liquid cooling loop 150. Meanwhile, by virtue of the serial connection of the server clusters, ΔT for the heating of the liquid in the cooling loop 150 is increased, as it represents the sum of each elevation of liquid temperatures of each server clusters 110, 112 . . . 11M. During operations, the cooling loop 150 alternates between transient thermal states and thermal steady state, where the cooling of the liquid in the cooling loop 150 (weighted-average of all the individual ΔTliquid) is equalled to ΔT.
Keeping these operational parameters in mind, as depicted in
Moreover, for the ALHEXs 120, 122 . . . 12M, the circulatory flow rate is the distributed global flow rate, which is reduced. As the circulatory flow rate in the ALHEXs is lower, the cold side pinch and the minimal temperature of the cooling liquid circulating within the liquid cooling loop 150 are reduced.
Thus, in operation, the liquid cooling arrangement 100 provides for the pump 130 to convey the cooling liquid to the serially-coupled server clusters 110, 112 . . . 11M via an increased local circulatory flow rate. In turn, the cooling liquid is conveyed, in a parallel manner, at an increased local circulatory flow rate in the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of each of the server clusters 110, 112 . . . 11M.
As the cooling liquid successively flows through the server clusters 110, 112 . . . 11M, it collects at least a portion of the thermal energy generated by the corresponding electronic equipment of each server cluster. However, the temperature of the cooling liquid remains within tolerable levels sufficient to continue to cool the electronic equipment until the cooling liquid is outputted by the last server cluster 11M. The liquid outputted by the last server cluster 11M is subsequently forwarded to the parallelized ALHEX 120, 122 . . . 12M, at a reduced global circulatory flow rate, for cooling the cooling liquid prior to being redirected to the server cluster 110 by the liquid back to pump 130.
In this manner, liquid cooling arrangement 100 provides a configuration that achieves a reduced global circulatory flow rate capable of efficiently cooling the electronic equipment contained within the server rack. As such, liquid cooling arrangement 100 allows for the use of different piping conduit configurations as well as pumps that are capable of effectively handling the reduced global circulatory flow rate, resulting in reduced costs of materials, equipment, and volume of cooling liquid required.
As shown, the liquid cooling arrangement 200 includes a plurality of server clusters 110, 112 . . . 11M and a plurality of ALHEXs 120, 122 . . . 12M that are fluidly connected in series to each other via the server rack liquid cooling loop 150. That is, each of the server clusters 110, 112 . . . 11M is serially coupled to a corresponding ALHEX 120, 122 . . . 12M via server rack liquid cooling loop 150. As noted above, liquid cooling loop 150 conveys and facilitates the circulatory flow of cooling liquid throughout the electronic equipment of the server rack. More specifically, internal fluid conduits of the ALHEX 120, 122 . . . 12M are fluidly connected in series with the server clusters 110, 112 . . . 11M, the liquid cooling units of a same server cluster being fluidly connected in parallel.
Each of the server clusters 110, 112 . . . 11M includes a plurality of server assemblies 110A-110N, 112A-112N . . . 11MA-11MN containing heat-generating electronic components, that are arranged in parallel fashion within the server clusters 110, 112 . . . 11M. In turn, each of the server assemblies 110A-110N, 112A-112N . . . 11MA-11MN incorporates at least one respective liquid cooling unit 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 for direct thermal contact liquid cooling of the heat-generating components, in which the respective liquid cooling unit 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 are accordingly arranged in a parallel manner.
The ALHEXs 120, 122 . . . 12M are configured to receive the heated cooling liquid from the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of the respective server clusters 110, 112 . . . 11M, via the server rack liquid cooling loop 150. The ALHEXs 120, 122 . . . 12M function to sufficiently air cool the heated liquid prior to forwarding the cooled cooling liquid to a succeeding server cluster 110, 112 . . . 11M.
The liquid cooling arrangement 200 additionally includes at least one pump 130 that is fluidly connected to the server rack liquid cooling loop 150. The pump 130 conveys the cooling liquid in the cooling loop 150, and thus causes the cooling liquid to flow from the last serially-coupled ALHEX 12M to the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of server clusters 110, 112 . . . 11M.
As such, liquid cooling arrangement 200 employs a configuration that combines serialized and parallelized components to achieve efficient circulatory flow rates and maintain adequate liquid cooling temperatures. In particular, by organizing the server clusters 110, 112 . . . 11M and ALHEXs 120, 122 . . . 12M in a serial configuration, the global circulatory flow rate of liquid cooling arrangement 200 is relatively reduced. In the meantime, by organizing the internal liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 in a parallel configuration, the global circulatory flow rate is relatively increased. Finally, by combination of such serialization and parallelization, the global circulatory flow rate is reduced.
However, in operation, liquid cooling arrangement 200 provides for the pump 130 to convey the cooling liquid between the serially-coupled server clusters 110, 112 . . . 11M and the ALHEXs 120, 122 . . . 12M with local increased circulatory flow rates. In turn, the cooling liquid flows in the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of each of the server clusters 110, 112 . . . 11M at an increased local circulatory flow rate.
As the cooling liquid successively flows through the server clusters 110, 112 . . . 11M, it collects at least a portion of the thermal energy generated by the corresponding electronic equipment of each server cluster. However, the cooling liquid is promptly cooled by a succeeding serially-coupled ALHEX 120, 122 . . . 12M that, in turn, forwards the air-cooled liquid to a subsequent server cluster 110, 112 . . . 11M. The air-cooled liquid outputted by the last ALHEX 12M is then directed back to pump 130 for recirculation at a reduced global circulatory flow rate.
The liquid cooling arrangement 200 therefore provides a configuration in which, due to the serialization of liquid cooled components, the global circulatory flow rate is reduced while the local circulatory flow rate within the server clusters 110, 112 . . . 11M, the internal liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1, and ALHEXs 120, 122 . . . 12M is increased.
In this manner, liquid cooling arrangement 200 achieves a reduced global circulatory flow rate capable of efficiently cooling the electronic equipment contained within the server rack. As such, liquid cooling arrangement 200 allows for the use of different piping conduit configurations as well as pumps that are capable of effectively handling the reduced global circulatory flow rate.
As noted above, each of the server clusters 110, 112 . . . 11M includes a plurality of server assemblies 110A-110N, 112A-112N . . . 11MA-11MN incorporating at least one respective liquid cooling unit 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 that are fluidly connected in parallel with one another within the server clusters 110, 112 . . . 11M.
Moreover, the ALHEXs 120, 122 . . . 12M are configured to receive the heated cooling liquid from the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of the respective server clusters 110, 112 . . . 11M, via the liquid cooling loop 150. The ALHEXs 120, 122 . . . 12M function to sufficiently air cool the heated liquid prior to forwarding the cooled cooling liquid to a succeeding server cluster 110, 112 . . . 11M.
The liquid cooling arrangement 300 additionally includes at least one pump 130 that is fluidly connected to the server rack liquid cooling loop 150. The pump 130 conveys the cooling liquid from the parallelized ALHEXs 120, 122 . . . 12M, via the server rack liquid cooling loop 150, and functions to forcibly provide the necessary circulatory flow rate of the cooling liquid through the server rack liquid cooling loop 150, in order to service the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of server clusters 110, 112 . . . 11M.
The liquid cooling arrangement 300 therefore provides a configuration in which the global circulatory flow rate is relatively reduced due to the serial connection of the server clusters 110, 112 . . . 11M with corresponding ALHEX 120, 122 . . . 12M. However, with the parallel connection of the server clusters 110, 112 . . . 11M and the parallel connection of the ALHEX 120, 122 . . . 12M, the global circulatory flow rate is increased. Finally, by combination of such serialization and parallelization, the global circulatory flow rate is increased. Moreover, with parallelization, the local circulatory flow rate within the server clusters 110, 112 . . . 11M and internal liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 is reduced.
In this manner, liquid cooling arrangement 300 achieves an increased global circulatory flow rate capable of efficiently cooling the electronic equipment contained within the server rack. As such, liquid cooling arrangement 300 allows for the use of different piping conduit configurations as well as pumps that are capable of effectively handling the increased global circulatory flow rate.
Each of the server clusters 110, 112 . . . 11M includes a plurality of server assemblies 110A-110N, 112A-112N . . . 11MA-11MN that are arranged in serial fashion within the server clusters 110, 112 . . . 11M. In turn, each of the server assemblies 110A-110N, 112A-112N . . . 11MA-11MN incorporates at least one respective liquid cooling unit 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 for direct thermal contact liquid cooling that are accordingly arranged in a serial manner. That is, in each server cluster 110, 112 . . . 11M, the cooling liquid is supplied to a liquid cooling unit 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 that forwards the cooling liquid to a subsequent serial liquid cooling unit, in which the last liquid cooling unit forwards the liquid to the serially-coupled ALHEX 120, 122 . . . 12M. Moreover, within each of the server clusters 110, 112 . . . 11M, the server assemblies 110A-110N, 112A-112N . . . 11MA-11MN that are arranged in series may comprise groups of server assemblies that are fluidly connected in a parallel manner.
The parallel ALHEXs 120, 122 . . . 12M are configured to receive the heated liquid from the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of the respective serially-coupled server clusters 110, 112 . . . 11M, via the server rack liquid cooling loop 150. The ALHEXs 120, 122 . . . 12M function to sufficiently air cool the heated cooling liquid prior to forwarding the cooled liquid to the succeeding server clusters 110, 112 . . . 11M.
The liquid cooling arrangement 400 additionally includes at least one pump 130 that is fluidly connected to the server rack liquid cooling loop 150. The pump 130 is configured to receive the cooling liquid from the parallelized ALHEXs 120, 122 . . . 12M, via the server rack liquid cooling loop 150, and functions to forcibly provide the necessary circulatory flow rate of the air-cooled liquid through the server rack liquid cooling loop 150 in a parallel manner, in order to service the liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 of server clusters 110, 112 . . . 11M.
Thus, in operation, liquid cooling arrangement 400 provides for pump 130 to supply air-cooled liquid to the parallelized combination of server clusters 110, 112 . . . 11M and serially-coupled ALHEXs 120, 122 . . . 12M via a reduced global circulatory flow rate. In turn, each of the server clusters 110, 112 . . . 11M supply air-cooled liquid to the serialized liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 at an increased local circulatory flow rate.
The liquid cooling arrangement 400 therefore provides a configuration in which the global circulatory flow rate is reduced due to the serial connection of the server clusters 110, 112 . . . 11M with corresponding ALHEX 120, 122 . . . 12M and the local circulatory flow rate within the server clusters 110, 112 . . . 11M is reduced while the local circulatory flow rate of the internal liquid cooling units 110A1-110N1, 112A1-112N1 . . . 11MA1-11MN1 is increased. Additionally, the local circulatory flow rate within the ALHEXs is also reduced.
In this manner, liquid cooling arrangement 400 provides a configuration that achieves a reduced global circulatory flow rate capable of efficiently cooling the electronic equipment contained within the server rack. As such, liquid cooling arrangement 400
Modifications and improvements to the above-described implementations of the present technology may become apparent to those skilled in the art. The foregoing description is intended to be exemplary rather than limiting. The scope of the present technology is therefore intended to be limited solely by the scope of the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
23305473.3 | Mar 2023 | EP | regional |