Maintenance of data center fluid-cooled devices

Abstract

A maintenance method for a liquid cooling system comprising a rack-mounted process assembly comprises a set of steps: a step of opening an intelligent control valve to a fully open state for a predetermined time period; and / or a step of operating at least one fan of the liquid cooling system at a maximum speed for a predetermined time period.

Description

[0001] Cross-referencing

[0002] This application claims priority to European Patent Application No. 24306317, filed on August 2, 2024, entitled “MAINTENANCE OF DATACENTERFLUID COOLING ARRANGEMENT”, the contents of which are incorporated herein by reference. Technical Field

[0003] This technology generally relates to the field of data center cooling system maintenance. Background Technology

[0004] Data centers and many computer processing facilities house numerous rack-mounted electronic processing units. During operation, these units generate significant amounts of heat, which must be dissipated to prevent electronic component failure and ensure continuous, efficient processing. Various air and / or liquid cooling systems have been implemented to facilitate the dissipation of heat generated by these electronic processing units.

[0005] As is well known, such systems typically include at least one liquid cooling circuit, which comprises multiple components such as pumps, heat exchangers, fans, and pipes for cooling liquid flow.

[0006] Data center maintenance, especially of its liquid cooling circuits, is crucial for ensuring continuous and efficient cooling of electronic processing equipment. However, maintenance can be extremely complex. In dense environments comprised of numerous server racks, operators must not only unclog and clean the pipes, but also the fans and heat exchangers.

[0007] That said, there remains interest in improving the maintenance of data centers that are air- and / or liquid-cooled.

[0008] The topics discussed in the background section should not be considered prior art simply because they are mentioned therein. Similarly, the problems mentioned in the background section, or those related to the topics in the background section, should not be considered as previously known in the prior art. The topics in the background section merely represent different approaches. Summary of the Invention

[0009] The present invention aims to improve this situation.

[0010] In one aspect, the inventive concept of this technology provides a method for maintaining a system including rack-mounted processing components, the system comprising: a dry cooling unit configured to supply coolant to the rack-mounted processing components and receive heat-reduced liquid from the rack-mounted processing components; a liquid distribution circuit configured to deliver coolant from the dry cooling unit to the rack-mounted processing components, the liquid distribution circuit including at least one heat exchanger (ALHEX) configured to utilize the coolant to cool an airflow through the rack; and at least one fan for cooling an airflow through the rack; wherein each of the rack-mounted data processing components includes: at least one heat-generating electronic processing element and at least one liquid cooling block arranged in relation to a corresponding heat-generating electronic processing element. The electronic processing element has thermal contact, the at least one liquid cooling block is fluidly coupled to the liquid distribution circuit to receive and circulate the cooling liquid through the liquid cooling block, and a smart control valve is correspondingly arranged to be fluidly coupled to the at least one liquid cooling block of the respective rack-mounted data processing component. The smart control valve is configured to control the flow rate of the cooling fluid of the respective rack-mounted data processing component based on detected temperature and pressure flow, independent of pressure, the flow rate including a range from a minimum flow rate when the valve is in a state called the closed state to a maximum flow rate when the valve is in a state called the fully open state; the method includes a set of steps referred to as maintenance steps: a step of opening the smart control valve to the fully open state for a predetermined time period, and / or a step of running the at least one fan at maximum speed for a predetermined time period.

[0011] This disclosure provides a reliable method for maintaining air and liquid cooling systems.

[0012] In some embodiments, the predetermined time in the step of opening the smart control valve to the fully open state is selected to be between 5 minutes and 30 minutes, preferably 15 minutes.

[0013] In some implementations, the predetermined time in the step of running the at least one fan at maximum speed for a predetermined time period is selected to be between 5 minutes and 30 minutes, preferably, the predetermined time in the step of running the at least one fan at maximum speed for a predetermined time period is selected to be 15 minutes.

[0014] In some embodiments, the method includes a step of stopping the fan before the step of running the at least one fan at maximum speed.

[0015] In some implementations, during normal use, the system is managed by a method referred to as a normal use method, which includes the following normal use steps:

[0016] - The current fluid flow rate and current input coolant temperature (T) of the corresponding rack-mounted data processing component. R-i ) and current output variable heat liquid temperature (T) R-o The measurement is performed, and the current temperature difference (ΔT) between the current input cooling liquid temperature and the current output heat transfer fluid temperature is calculated.

[0017] - The current temperature difference is compared with the target temperature value, and the liquid flow rate of the intelligent control valve is dynamically adjusted based on the comparison result, the current liquid flow rate, and the current input cooling liquid temperature.

[0018] - The internal temperature (T) of the rack-mounted processing assembly chip The measurement is performed, and, if the current temperature difference equals the target temperature value,

[0019] - Determine the current input coolant temperature (T) R-i The internal temperature (T) of the rack-mounted processing assembly described below chip Whether it is less than the predetermined limit value, and

[0020] - The internal temperature (T) of at least one air-cooled electronic processing component air cooled IT Measurements were taken of the internal temperature (T) of the rack-mounted processing assembly. chip If the value is less than the predetermined limit, and

[0021] - Determine the current input coolant temperature (T) R-i Under these conditions, the internal temperature (T) of the at least one air-cooled electronic processing element air cooled IT Is it less than the predetermined limit value?

[0022] The maintenance method includes the following steps: before opening the intelligent control valve to the fully open state, at least stop the step of dynamically adjusting the liquid flow rate of the intelligent control valve based on the comparison result, the current liquid flow rate, and the current input cooling liquid temperature; and after the step of opening the intelligent control valve to the fully open state is completed, perform the step of dynamically adjusting the liquid flow rate of the intelligent control valve based on the comparison result, the current liquid flow rate, and the current input cooling liquid temperature.

[0023] In some implementations, the method includes sending notifications and / or alarms after the maintenance step, before the maintenance step, or during the maintenance step.

[0024] In some embodiments, the method further includes: a step of measuring at least one set of parameters prior to the maintenance step, the measured values ​​of the set prior to the maintenance step being referred to as an initial maintenance set; and a step of measuring the at least one set of parameters after the maintenance step, the measured values ​​of the set after the maintenance step being referred to as a final maintenance set.

[0025] In some implementations, the method includes the step of recording each measured value of the parameter.

[0026] In some implementations, the method includes evaluating the maintenance steps based on the final maintenance group or based on the difference between the final maintenance group and the initial maintenance group.

[0027] In some embodiments, the at least one parameter is selected as the air temperature or the temperature of the heating element at the outlet of the rack, or the input cooling liquid temperature (T). R-i The corresponding output temperature of the rack-mounted data processing component is the variable heat liquid temperature (T). R-o ), the internal temperature (T) of the rack-mounted processing assembly chip ), or the temperature of air-cooled electronic processing components (T) air cooled IT The air temperature downstream of the heat exchanger (T) air-c The air temperature upstream of the heat exchanger (T) air-h ).

[0028] The present invention also relates to a computer program comprising instructions that, when executed by a computer, cause the computer to perform steps of the maintenance method already described.

[0029] Advantageously, if the current input coolant temperature (T) R-i Under these conditions, the internal temperature (T) of the at least one air-cooled electronic processing elementair cooled IT If the speed exceeds the predetermined limit, the fan speed will be increased.

[0030] Advantageously, the method further includes: after the fan speed increases, determining the current input coolant temperature (T). R-i The internal temperature (T) of at least one air-cooled electronic processing element is below air cooled IT Is it less than the predetermined limit value?

[0031] Advantageously, if the increased fan speed is the maximum speed, an indication is issued that the fan speed has reached the maximum speed.

[0032] Advantageously, the method further includes: controlling the temperature (T) of the hot air stream before it passes through at least one heat exchanger. air-h Measurements were taken, and the internal temperature (T) of the at least one air-cooled electronic processing element was measured. air cooled IT If the current input coolant temperature (T) is less than the predetermined limit, determine if the current input coolant temperature (T) is within the predetermined limit. R-i Under these conditions, the temperature of the hot air stream (T) air-h Is it less than the predetermined limit value?

[0033] Advantageously, when the temperature of the hot air stream (T) air-h When the value exceeds a predetermined limit, the method includes increasing the fan speed.

[0034] Advantageously, the method includes: after the fan speed increases, determining the current input coolant temperature (T). R-i ) Temperature of hot airflow (T) air-h Is it less than the predetermined limit value?

[0035] Advantageously, if the increased fan speed is the maximum speed, an indication is issued that the fan speed has reached the maximum speed.

[0036] Advantageously, the method also includes: measuring the temperature (T) of the cold air stream after it has passed through at least one heat exchanger. air-c ) and measure the temperature of the cold air flow (T) air-c ) and the current input coolant temperature (T) R-i The difference between (Pin) c The temperature of the cold air flow (T) is determined. air-c ) and the current input coolant temperature (T) R-i The difference between (Pin) c This is called the cooling difference; and, at the temperature of the hot air flow (T) air-h If the current input coolant temperature (T) is less than the predetermined limit, determine if the current input coolant temperature (T) is within the predetermined limit. R-i The following describes the cooling difference (Pin)c Is it less than the predetermined limit value?

[0037] Advantageously, when the cooling difference (Pin) c When the value exceeds a predetermined limit, the method includes reducing the fan speed.

[0038] Advantageously, the method includes: after the fan speed decreases, determining the current input coolant temperature (T). R-i Under these conditions, the cooling difference (Pin) c Is it less than the predetermined limit value?

[0039] Advantageously, if the reduced fan speed is the minimum speed, an indication is issued that the fan speed has reached the minimum speed.

[0040] Advantageously, the method also includes increasing the liquid flow rate of the corresponding intelligent control valve when the current temperature difference is greater than the target temperature difference.

[0041] Advantageously, the method further includes: when the current temperature difference is less than the target temperature difference, after confirming that the reduced flow rate is not lower than the minimum flow limit, reducing the liquid flow rate of the corresponding smart control valve.

[0042] In a related aspect of the inventive concept, this technology provides a fluid cooling system for rack-mounted processing assemblies, the fluid cooling system comprising: a liquid cooling facility for supplying cooling liquid to the rack-mounted processing assembly and receiving heat-reduced liquid from the rack-mounted processing assembly; a liquid distribution circuit for distributing cooling liquid from the liquid cooling facility to the rack-mounted processing assembly, the liquid distribution circuit including at least one heat exchanger (ALHEX) configured to utilize the cooling liquid to cool the airflow of the rack; wherein each of the rack-mounted data processing assemblies includes: at least one heat-generating electronic processing element and at least one liquid cooling block. The at least one liquid cooling block is arranged to be in thermal contact with a corresponding at least one heat-generating electronic processing element, the at least one liquid cooling block being fluidly coupled to the liquid distribution circuit to receive the cooling liquid and circulate the cooling liquid through the liquid cooling block, and a smart control valve is correspondingly arranged to be fluidly coupled to the at least one liquid cooling block of the corresponding rack-mounted data processing component, the smart control valve being configured to control the flow rate of the cooling fluid of the corresponding rack-mounted data processing component based on the detected temperature and the monitored flow rate, regardless of pressure; wherein the at least one electronic processing element is air-cooled by at least one fan; wherein the system is configured to operate the method already described.

[0043] The present invention also relates to a computer program comprising instructions that, when executed by a computer, cause the computer to perform the steps of a described method.

[0044] The present invention relates to a leak detection system, and more particularly, the leak detection system is used for a fluid cooling system of a rack-mounted processing assembly, the leak detection system including electrical circuitry configured to disconnect under normal operating conditions of the fluid cooling system and to close in the event of a liquid leak.

[0045] The present invention also relates to a computer-readable medium comprising instructions that, when executed by a computer, cause the computer to perform the steps of the maintenance method already described.

[0046] In the context of this specification, unless otherwise expressly stated, computer system may refer to, but is not limited to, “electronic device,” “operating system,” “system,” “computer-based system,” “controller unit,” “monitoring device,” “control device,” and / or any combination thereof suitable for the present and relevant task.

[0047] In the context of this specification, unless otherwise expressly stated, the terms "computer-readable medium" and "memory" are intended to include media of any nature and kind, with non-limiting examples including RAM, ROM, disks (CD-ROM, DVD, floppy disk, hard disk, etc.), USB keys, flash memory cards, solid-state drives, and tape drives. Still within the context of this specification, "a" computer-readable medium and "the" computer-readable medium should not be construed as the same computer-readable medium. Rather, and where appropriate, "a" computer-readable medium and "the" computer-readable medium may also be construed as a first computer-readable medium and a second computer-readable medium.

[0048] In the context of this specification, unless otherwise expressly stated, the words “first,” “second,” “third,” etc., are used as adjectives only for the purpose of distinguishing the nouns they modify from one another, and not to describe any particular relationship between these nouns.

[0049] Each implementation of this technology has at least one of the above-mentioned objectives and / or aspects, but not necessarily all of them. It should be understood that some aspects of this technology arising from attempts to achieve the above objectives may not satisfy those objectives and / or may satisfy other objectives not specifically listed herein.

[0050] Additional and / or alternative features, aspects, and advantages of the present technology will become apparent from the following description, the accompanying drawings, and the appended claims. Attached Figure Description

[0051] To better understand this technology, as well as other aspects and additional features thereof, reference is made to the following description used in conjunction with the accompanying drawings, in which:

[0052] Figure 1 A functional block diagram depicts an internal server cluster configuration of a data center server rack according to a non-limiting embodiment of the present disclosure;

[0053] Figure 2 Some non-limiting embodiments according to this disclosure are described. Figure 1 Functional block diagram of the system's cooling device;

[0054] Figure 3 The present disclosure describes the equipment equipped with some non-limiting embodiments. Figure 2 A schematic physical view of the cooling unit's frame, showing the location of the sensors;

[0055] Figure 4 A flowchart is shown of a continuous operation process for controlling and balancing a liquid cooling system for a rack-mounted processing assembly according to a non-limiting embodiment of the present technology.

[0056] Figures 5 to 12 They are shown respectively Figure 4 The flowchart of the corresponding subroutine of the process;

[0057] Figures 13 to 15 Flowcharts of maintenance methods according to some non-limiting embodiments of the present technology are shown respectively;

[0058] Figure 16 This allows for the implementation of some non-limiting methods based on the present technology. Figures 13 to 15 A block diagram of the maintenance method module.

[0059] It should be understood that in the accompanying drawings and corresponding descriptions, the same features are represented by the same reference numerals. Furthermore, it should be understood that the drawings and following description are intended for illustrative purposes only, and their disclosure is not intended to limit the scope of the claims. Detailed Implementation

[0060] This disclosure aims to address at least some of the problems associated with the conventional use of various piping configurations and numerous pumps to supply liquid flow to water blocks to meet the cooling needs of numerous heat-generating components. Specifically, this disclosure proposes embodiments of air and liquid cooling systems that are prioritized for servers with high heat generation to optimize their cooling performance while simplifying the arrangement of liquid cooling piping.

[0061] The examples and conditional language listed herein are primarily intended to help the reader understand the principles of this technology, and are not intended to limit the scope of this technology to these specific examples and conditions. It will be understood that those skilled in the art can design various arrangements and structures that, while not explicitly described or shown herein, still embody the principles of this technology.

[0062] Furthermore, to aid understanding, the following description illustrates a relatively simplified implementation of this technology. As those skilled in the art will understand, various implementations of this technology may involve greater complexity.

[0063] In some cases, examples that are considered helpful in modifying the present technology may also be illustrated. This is done merely to aid understanding and is not intended to limit the scope or boundaries of the present 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 present technology. Furthermore, the absence of examples of modifications should not be construed as impossibility of modification or as the only way to implement that element of the present technology.

[0064] Furthermore, all statements herein that illustrate the principles, aspects, and implementations of the present technology, as well as specific examples of the present technology, are intended to cover structural and functional equivalents of the present technology, whether such structural and functional equivalents are currently known or will be developed in the future. Therefore, for example, those skilled in the art will understand that any block diagram herein represents a conceptual view of an illustrative system embodying the principles of the present technology.

[0065] Based on the above, we will now consider some non-limiting examples to illustrate implementations of various inventive aspects of this disclosure.

[0066] Figure 1 A functional block diagram depicts a representative internal server cluster 10 configuration of a data center server rack according to a non-limiting embodiment of the present disclosure. This representative server cluster 10 configuration includes multiple server sub-clusters 20-2M, each of which correspondingly includes multiple data processing components 20A-20N to 2MA-2MN, each data processing component containing heat-generating electronic processing elements.

[0067] As shown, each of the data processing components 20A-20N to 2MA-2MN includes at least one corresponding liquid cooling block unit 20A1-20N1 to 2MA1-2MN1, which is configured to be in direct thermal contact with the heat-generating electronic processing components. Each of the liquid cooling block units 20A1-20N1 to 2MA1-2MN1 is equipped with an internal conduit (not shown) for containing a channel-type cooling liquid that circulates through the liquid cooling block unit. The channel-type cooling liquid is supplied by a cooling liquid supply unit, which sequentially delivers it to each of the liquid cooling block units 20A1-20N1 to 2MA1-2MN1 through the internal server cluster liquid circulation channel 30 to absorb heat energy from the heat-generating electronic components and discharge the heated liquid from the liquid cooling block unit.

[0068] Other components in the data processing units 20A-20N to 2MA-2MN that generate less heat can be air-cooled by the fan of the device 100. Heat-generating components include, for example, the graphics processing unit (GPU) and / or the central processing unit (CPU). Other components include, for example, random access memory (RAM), hard disk drives, etc.

[0069] Given the configuration of the aforementioned internal data center server cluster 10, Figure 2 A functional block diagram of a fluid cooling device 100 according to a non-limiting embodiment of this disclosure is depicted. While the embodiments and implementations described herein are directed to data center racks, it is understood that extending such concepts to cover multiple data center racks is clearly within the scope of this technology.

[0070] As shown, the fluid cooling device 100 includes a single liquid distribution loop 105 configured with a supply side and a return side. The supply side supplies coolant from a liquid cooling facility 170 to the server clusters 130-13M, 140-14P, and 150-15L in the rack. The return side returns heated liquid from the server clusters to the liquid cooling facility 170 for recooling and recirculation back to the server clusters 130-15L. (As mentioned above...) Figure 1 As described, each server cluster in server clusters 130-15L includes a data processing component (e.g., see...). Figure 1 The data processing component (20A-20N to 2MA-2MN) includes at least one corresponding liquid cooling block unit for direct thermal contact with the heat-generating electronic components of the data processing component, thereby dissipating heat therein. The liquid cooling block is also called a water block (…). Figure 3 WB in the middle.

[0071] The liquid distribution loop 105 is configured with: a liquid distribution inlet 101 along the supply side for supplying coolant to the fluidly connected server cluster 130-15L; and a liquid distribution outlet 102 along the return side for receiving heated liquid from the server cluster 130-15L and returning the heated liquid to the liquid cooling facility 170 for recooling and recirculation back to the server cluster 130-15L. The liquid distribution loop 105 may be constructed of flexible materials (e.g., rubber, plastic, etc.), rigid materials (e.g., metal, PVC pipes, etc.), or any combination thereof. It is understood that the delivered liquid may include water, alcohol, or any suitable liquid capable of maintaining a sufficient cooling temperature.

[0072] The liquid cooling facility 170 may include a dry cooler unit 172 configured to process and reconditioned liquid received from the server rack to provide cooling liquid, which is then recirculated back to the server cluster 130-15L via a liquid distribution loop 105. The liquid cooling facility 170 may also include a pump 174 configured to provide the required pressure increments and volumetric flow rates of the cooling liquid from the dry cooler unit 172 throughout the liquid distribution loop 105.

[0073] The fluid-cooled device 100 also includes a plurality of air-liquid heat exchangers (ALHEX) 110-114. In the illustrated embodiment, the ALHEX 110-114 are fluidly connected in parallel via a liquid cooling circuit 105 and are also fluidly connected to the server cluster 130-15L via the liquid cooling circuit 105. However, it is understood that the ALHEX 110-114 can be fluidly interconnected in other configurations, such as in series via the liquid cooling circuit 105, without departing from the concept of the disclosed art.

[0074] The ALHEX 110-114 is used to adequately cool the ambient air surrounding the server cluster 130-15L. The ALHEX 110-114 can be configured in any suitable way that reduces the liquid temperature through the supplied airflow (e.g., via a compact fan), such as internal cooling coils, heat dissipation airflow fins, etc. The ALHEX 110-114 can be positioned, for example, at the rear of the rack housing the server cluster 130-15L to directly cool the air leaving the server cluster 130-15L and warmed by the air-cooling components therein.

[0075] The temperature change of the water in distribution loop 105 can be described as follows. Water flows out from liquid cooling facility 170 and enters distribution loop 105 at a "cold" temperature. It continuously heats up in loop 105, first in heat exchangers ALHEX 110-114, and then into server clusters 130-15L. After the "warm" liquid circulates within each of the data processing components and cooling units in server clusters 130-15L, the liquid flowing out from standard priority server clusters 130-15L is heated. The "heated" liquid is transferred to the return side of liquid distribution loop 105 to return to liquid cooling facility 170 for recooling and recirculation back to all server clusters 120-15L. In some implementations, the temperature of the heated liquid can be in the range of approximately 45°C to 65°C, while the "cold" temperature is selected between 20°C and 40°C.

[0076] like Figure 3 As shown, the fluid cooling device 100 includes multiple sensors. The device 100 includes a liquid temperature sensor LTS, which is preferably mounted as follows:

[0077] - Installed at the inlet of each rack, the measured temperature is the inlet liquid temperature T. R-i (or "T") in ”);

[0078] - Installed at the outlet of each rack, the measured temperature is the outlet liquid temperature T. R-o (or "T") out ”).

[0079] The device 100 also includes an air temperature sensor (ATS), preferably installed upstream and downstream of each heat exchanger ALHEX 110-114. For each ALHEX j (j is 110, 112, or 114), the measured temperature is denoted as T. air-h-j and T air-c-j , where h means hot and c means cold.

[0080] The device 100 also preferably includes at least one ( Figure 3 There are three air temperature sensors (ATS), at least one of which is used to measure the air temperature on each rack. air-R-110 T air-R-112 T air-R-114 More generally written as T air-R-i .

[0081] like Figure 3As shown, each rack-mounted server cluster in the rack-mounted server clusters 130-15L is fluidly connected to a “smart” valve SV, which dynamically controls the liquid flow rate of the corresponding processing unit 130-15L based on detected liquid temperature. For the purposes of this disclosure, the term “smart” valve refers to a pressure-independent, temperature-responsive valve that includes a differential pressure regulator to automatically regulate changes in system pressure. Such smart valves may include PICVs (“pressure-independent control valves”), such as valves with a combination of the registered trademark AB-QM and an actuator, or other functionally similar valves, or combinations of these valves, such as a combination of a solenoid valve and a control valve. The device 100 preferably includes a rack-mounted flow meter sensor FS for measuring the liquid flow rate (denoted as V) of the smart valve. The flow rate varies from a minimum flow rate (preferably zero) in the closed state of the smart valve to a maximum flow rate in the fully open state of the smart valve.

[0082] Device 100 includes at least one fan. Figure 3 In this device 100, there are corresponding fans dedicated to each heat exchanger 110, 112, 114. The fan speed is expressed in RPM (revolutions per minute), and %RPM means the percentage of the maximum RPM of the fan model under consideration.

[0083] As will be described in detail below, T is based on measurements taken from each server cluster in server cluster 130-15L. R-i and T R-o The corresponding intelligent valves are used to dynamically control the individual liquid flow rate of each rack-mounted processing unit in the rack-mounted processing assembly to balance and maintain an optimal target temperature difference ΔT between the returned heat-generating liquid and the supplied cooling / recooling liquid of system 100. Maintaining this optimal temperature difference ΔT improves the overall cooling system efficiency. In practical applications, the temperature difference ΔT is a positive value.

[0084] The temperature of the heat-generating electronic processing element is called the processing component temperature, T. chip The heat-generating electronic processing element is cooled by a liquid cooling block arranged in thermal contact with the corresponding electronic processing element.

[0085] The temperature of the air-cooled element is denoted as T. air cooled IT .

[0086] As will be described in detail below, another parameter of note is the "pinch" value of ALHEX 110, 112, and 114. That is, each of the ALHEX 110, 112, and 114 has a "hot side" and a "cold side." For the hot side, the pinch value ΔT hotpinchThe pinch value is defined as the temperature difference between the hot air entering the ALHEX for cooling and the temperature of the hot liquid leaving the ALHEX. For the cold side, the pinch value is ΔT. coldpinch Pinch value ΔT is defined as the temperature difference between the temperature of the cooling air leaving ALHEX and the temperature of the cold liquid entering ALHEX. hotpinch and ΔT coldpinch Both are positive numbers. From now on, the pinch value on the cold side will be denoted as Pin. c .

[0087] The following description now focuses on the operation method 250 of system 100, that is, under the normal operating conditions of system 100.

[0088] from Figure 3 As can be seen, device 100 includes a leak sensor for detecting liquid leaks in loop 105. The sensor uses two wires: one connected to ground and the other to a positive voltage source (+), typically placed near the floor or in any area where a leak may occur. The positive voltage is typically 3.3V, 5V, or 12V. When water comes into contact with the two wires, current flows, thus detecting a liquid leak. The sensor generates a signal that can be used to trigger an alarm or send a notification to the compartment controller.

[0089] As described above, maintaining an optimal temperature difference ΔT results in improved overall cooling system efficiency. Method 250 according to this disclosure aims to maintain ΔT at a given value while taking into account the coolant flow rate, the temperatures of the cold and hot air, the temperatures of the coolant and hot liquid, and the fan speed. In other words, the parameters of method 250 are ΔT, V, and T. R-i T R-o T air-c T air-h Pin c T chip T air cooled IT .

[0090] Figure 4 A flowchart illustrating an operation process 250 of a liquid-cooled device 100 for controlling and balancing rack-mounted processing units according to a non-limiting embodiment of the present technology is shown. In some embodiments, the operation process 250, or a portion thereof, is performed by individual rack-mounted processing units, each directly connected to a smart valve. In other embodiments, the operation process 250, or a portion thereof, may be performed by a main control unit communicatively coupled to each of the individual rack-mounted processing units. For the purposes of this disclosure, it is not intended that one or more specific entities perform the initialization process in accordance with the concepts provided herein.

[0091] Operation 250 begins with task block 252, in which, for each individual rack-mounted server cluster 130-15L, the liquid flow rate V of the rack-mounted component and the temperature T of the heat-transferring liquid flowing out of the rack-mounted component are monitored. out And the temperature T of the 130-15L coolant entering the rack-mounted server cluster. in Take measurements.

[0092] Then process 250 proceeds to decision block 254, where it is determined whether the temperature difference ΔT between the outflowing heat-generating liquid and the inflowing cooling liquid is negative to the target temperature difference XK (within the tolerance range ± ZK). If the temperature difference ΔT between the outflowing heat-generating liquid and the inflowing cooling liquid is not lower than the target temperature difference XK (within the tolerance range ± ZK), then decision block 256 determines whether the temperature difference ΔT is greater than the allowable target temperature difference XK ± ZK. If the temperature difference ΔT is greater than the allowable target temperature difference XK ± ZK, then in task block 260, the liquid flow rate V is increased by a predetermined value. Then, process 250 proceeds... Figure 5 The subroutine (SR) 262 in the diagram determines whether the smart valve (referred to as PICV in the diagram) is fully open; if the smart valve is fully open, it determines the increased liquid flow rate V. n Is it equal to the predetermined maximum flow rate V? max .

[0093] As can be seen from the diagram, if the smart valve is not fully opened (block 5-1), process 250 returns to decision block 256.

[0094] If the smart valve is fully open (the position sensor detects that the smart valve is fully open), a warning is issued (5-2), and then at step 5-3 the liquid flow rate is adjusted to the maximum flow rate (V). max of PICV ) for comparison. If V n If the flow rate is less than the maximum flow rate (5-4), an alarm will be issued: insufficient liquid flow, and process 250 will issue an alarm message (5-5) and exit the process (e.g.). Figure 4 (See steps 264-266-268 in the text). If V n Greater than the maximum liquid flow rate V max If so, an alarm (5-6) will be issued: the sensor or smart valve may be malfunctioning and assistance is needed, and process 250 will exit the process (e.g., Figure 4 (See steps 264-266-268 in the text).

[0095] In other words, when the smart valve is fully open, if the measured flow rate corresponds to the maximum flow rate, only a warning is issued; if the measured flow rate does not correspond to the level that the flow rate should reach, an alarm is issued.

[0096] Returning to decision block 256, if it is determined that the temperature difference ΔT is not greater than the allowable target temperature difference XK±ZK (an example of X=20 and Z=1 is shown in the figure), then process 250 reduces the liquid flow rate by a predetermined value at task block 270, and then determines the reduced liquid flow rate V at decision block 272. n+1 Is it less than the predetermined minimum liquid flow rate V? min Minimum fluid flow rate V min It is configured to increase laminar flow within the liquid circuit 105.

[0097] If decision block 272 determines V n+1 Not less than V min Then process 250 returns to task block 252 to process the V and T values ​​of the rack-mounted component. in and T out Perform a remeasurement. If decision block 272 determines V... n+1 Less than V min Then process 250 proceeds to task block 274 to increase the liquid flow rate V by a predetermined value; and issues a notification indicating that the target temperature difference ΔT is underloaded or that the sensor or smart valve may be malfunctioning.

[0098] Returning to decision block 254, if it is determined that the temperature difference ΔT equals the target temperature difference XK (within the tolerance range ±ZK), then process 250 proceeds to task block 276 to determine the temperature T of the coolant. R-i (T in The internal temperature parameters (i.e., processing unit temperature T) of certain rack-mounted processing components in a 130-15L server cluster. chip Is it lower than the predetermined high temperature limit? Figure 6 As shown in detail. Each chip can have appropriate limit values, or some or all chips can have the same limit values.

[0099] As can be seen from the figure, at step 6-1, at T R-i Next, T chip The temperature is compared to the limit value. If decision block 6-1 determines the processing unit temperature T... chip If the value exceeds the limit, process 250 proceeds to decision block 6-2 to analyze whether the number of affected servers exceeds a predetermined value (e.g., 20%). If the number exceeds the predetermined value, process 250 issues a statement regarding the chip temperature T. chipAlarm 6-3 is issued, warning that servers exceeding a predetermined value are affected. The subroutine then exits to task block 278. If the quantity is less than the predetermined value, the process issues an alarm to check the tightness of the chip water block and the application of 2 (the "thermal interface material (TIM)" located between the processor and the water block) on the affected server at step 6-4. Then, process 250 exits the subroutine and proceeds to decision block 280. If T at decision block 6-1... chip If the value is not greater than the limit, process 250 exits the subroutine and proceeds to decision block 280.

[0100] Subroutine 278 will be explained in detail later with reference to subroutine 292, such as... Figure 10 As shown.

[0101] Proceeding to decision block 280, process 250 determines the appropriate cooling liquid T. R-i In T R-i Below, other internal temperature parameters of the rack-mounted processing unit 130A-150N (i.e., airflow temperature T) air cooled IT Whether it is less than or equal to the predetermined high temperature limit value.

[0102] If not, then process 250 starts subroutine 282, such as... Figure 7 As shown. In task block 7-1, the fan speed is increased by a predetermined value (e.g., 5%). Then, in step 7-2, it is checked whether the fan speed has reached the maximum speed (100%). If the fan speed has reached the maximum speed, a warning is issued in step 7-3 indicating that the fan has reached its maximum speed. Then, in step 7-4, the air cooling component temperature T is checked again. air cooled IT The value is compared with the limit. If the limit is not reached, process 250 exits the subroutine and proceeds to decision block 284. If the limit is reached, process 250 issues an alarm at step 7-5, T. air cooled IT If the value exceeds the limit and the fan has reached its maximum speed, exit the subroutine.

[0103] Returning to step 7-2, if the fan speed is still below the maximum speed, then at step 7-6, again set the air-cooled component temperature T. air cooled IT The process is compared with the limit value. If the limit value is not reached, process 250 exits the subroutine and proceeds to decision block 284; if the limit value is reached, the process returns to step 7-1.

[0104] Proceeding to decision block 284, process 250 determines: for the temperature T of the coolant... R-i The temperature T of the hot air stream entering ALHEX air-h Is it less than or equal to the predetermined high temperature limit?

[0105] If not, then process 250 starts subroutine 286, such as... Figure 8 As shown. At task block 8-1, the fan speed is increased by a predetermined value (e.g., 5%). Then, at step 8-2, it is checked whether the fan speed has reached the maximum speed (100%). If the fan speed has reached the maximum speed, a warning is issued at step 8-3 indicating that the fan has reached its maximum speed. Then, at step 8-4, the hot airflow temperature T is again... air-h The value is compared with the limit. If the limit is not reached, process 250 exits the subroutine and proceeds to decision block 288. If the limit is reached, process 250 issues an alarm at step 8-5, T. air-h The value is greater than the limit and the fan has reached its maximum speed, and process 250 exits the subroutine and proceeds to decision block 288.

[0106] Returning to step 8-2, if the fan speed is still below the maximum speed, then at step 8-6, again adjust the airflow temperature T. air-h The process is compared with the limit value. If the limit value is not reached, process 250 exits the subroutine and proceeds to decision block 288; however, if the limit value is reached, process 250 returns to step 8-1.

[0107] Returning to decision block 284, if the temperature of the hot air stream is below the limit value, process 250 proceeds to decision block 288.

[0108] Returning to decision block 280, if the air-cooled component temperature T air cooled IT If the temperature is below the upper limit, then at decision block 290, the hot air flow temperature T will be adjusted. air-h Compare with the limit value (in T) R-i (Next). If the limit is reached, process 250 initiates the interpreted subroutine 286.

[0109] Proceeding to decision block 288, process 250 determines: for the temperature T of the coolant... in Cold difference Pin c Is it less than or equal to the predetermined upper limit of the cooling difference? If not, then process 250 starts subroutine 292, such as... Figure 10 As shown.

[0110] Returning to decision block 290, if the limit value is not reached, then at decision block 294, the cold difference pin will be adjusted. c The value is compared with a predetermined limit. If the limit is reached, process 250 initiates subroutine 296, as follows. Figure 9 As shown.

[0111] like Figure 9As can be seen, at step 9-1, the fan speed is reduced by a predetermined value (e.g., 5%). Then, at step 9-2, it is checked whether the fan speed has reached the minimum speed (e.g., 30%). If the fan speed has reached the minimum speed, a warning is issued at step 9-3 that the fan has reached its minimum speed, and at step 9-4, the fan speed is increased by a predetermined value (e.g., 5%) to ensure safety. Then, process 250 exits the subroutine and proceeds to decision block 298. Returning to step 9-2, if the fan speed is greater than the minimum speed, at step 9-5, the cooling difference pin is adjusted. c With limit value (in T) R-i The following steps are compared. If the limit value is not reached, process 250 returns to step 9-1. If the limit value is reached, process 250 executes step 9-4 and exits to decision block 298.

[0112] At decision block 298, the cold difference pin is adjusted again. c Compare with the limit value. If the limit value is reached, process 250 starts subroutine 292, such as... Figure 10 As shown, it will be described below.

[0113] exist Figure 10 In this context, the subroutine may be applicable to processing components at temperatures T. chip (278), or applicable to Pin c Cold temperature difference (292).

[0114] As can be seen from the diagram, in step 10-1, the liquid flow rate is increased by a predetermined value. If the smart valve is fully open, a warning is issued (10-2).

[0115] In step 10-3, the liquid flow rate is compared with the maximum flow rate (V). max of PICV ) for comparison. If V n If the flow rate is less than the maximum flow rate (10⁻⁴), an alarm will be issued: Insufficient liquid flow, and an alarm message (10⁻⁵) will be sent. If V n If the flow rate exceeds the maximum, an alarm (10-6) will be issued: there may be a malfunction in the sensor or smart valve function, assistance is needed, and the process will be terminated.

[0116] After steps 10-3, 10-5, and 10-6, at step 10-7, the temperature T is... x (For subroutine 278, T) x For T chip For subroutine 292, T x Pin c ) compared with the limit value (in T) R-i (Next). If the limit value is not reached, check the temperature difference ΔT at step 10-8. If the limit value is reached, set T at step 10-9.x With T chip Compare. If T x =T chip (That is, if the subroutine being executed is 278), then an alarm (10-10) is issued, and the temperature T... chip If it exceeds its limit, check the temperature difference ΔT at step 10-8. Return to step 10-9; if T... x With T chip If the subroutine is different (i.e., the subroutine being executed is 292), then a warning (10-11) is issued, Pin c The process is high, and process 250 continues with steps 10-8.

[0117] Returning to step 10-1, if the smart valve is not fully open, then at T R-i At that time, T x Compare it with its limit value (10-12). If the limit value is reached, process 250 returns to step 10-1. If the limit value is not reached, check the temperature difference ΔT at step 10-8.

[0118] At step 10-8, if the temperature difference ΔT is less than the permissible target temperature difference XK±ZK, a warning (10-13) is issued: the temperature difference ΔT is less than the target temperature difference, and process 250 exits subroutine 292 / 278. At step 10-8, if the temperature difference ΔT is not less than the permissible target temperature difference XK±ZK, then process 250 exits subroutine 292 / 278.

[0119] When exiting subroutine 292, process 250 starts subroutine 304.

[0120] Returning to decision block 294, if the cold difference Pin c If the value is less than the predetermined limit, process 250 initiates subroutine 300 to perform a final check on the fan. Figure 11 As shown.

[0121] like Figure 11 As can be seen, at task block 11-1, the fan speed is reduced by a predetermined value (e.g., 5%) but not below a predetermined value (e.g., 30%).

[0122] Then, at step 11-2, at T R-i Below, the temperature T air cooled IT The value is compared with the limit. If the limit is reached, the fan speed is increased by a predetermined value (e.g., 5%) at step 11-3, and process 250 exits subroutine 300. Returning to step 11-2, if the limit is not reached, at step 11-4, the fan speed is increased by a predetermined value (e.g., 5%). R-i Below, the temperature T air-hThe value is compared with the limit. If the limit is reached, the fan speed is increased by a predetermined value (e.g., 5%) at step 11-3, and process 250 exits subroutine 300. Returning to step 11-4, if the limit is not reached, at step 11-5, the fan speed is increased by a predetermined value (e.g., 5%). R-i Below, the cooling difference Pin c The value is compared with a limit. If the limit is reached, the fan speed is increased by a predetermined value (e.g., 5%) at step 11-3, and program 250 exits subroutine 300. If the limit is not reached at step 11-5, subroutine 300 returns to step 11-1.

[0123] When subroutine 300 completes, process 250 initiates subroutine 302 to perform a final check on the liquid flow rate, such as... Figure 12 As shown.

[0124] As can be seen from the figure, process 250 reduces the liquid flow rate by a predetermined value at task block 12-1, and then at step 12-2, it determines whether the reduced liquid flow rate is less than or equal to the predetermined minimum liquid flow rate V. min If V n If the value is greater than the minimum, then at step 12-3, the temperature T will be... chip Compare with its limit value (in T) R-i If the limit value is not reached, then in step 12-4, the temperature difference ΔT is compared with a target value that is higher than the previous target value X. Figure 12 An example is shown with a target value of 25K while considering X equal to 20. If this limit is not reached, then at step 12-5, at T... R-i Below, the temperature T air cooled IT Compare with the limit value. If the limit value is not met, then at steps 12-6, adjust the cooling difference (Pin). c Compare with its limit value (per pin) c (The same limit applies). If the limit is not met, then at step 12-7, at T R-i Next, T air-h It is compared with its limit value. If the limit value is not reached, execution process 250 returns to step 12-1.

[0125] At each of steps 12-2, 12-3, 12-4, 12-5, 12-6, and 12-7, if the limit value is reached, the liquid flow rate is increased by a predetermined value at task block 12-8, and process 250 exits subroutine 302.

[0126] After subroutine 302 completes, process 250 initiates subroutine 304 to perform a final review of the value. In subroutine 304, the value to be published, namely the liquid flow rate (m³), is recorded.3 / h), temperature difference (K), temperature (K) as a parameter of process 250, smart valve opening (%), fan speed (%), pin c (K), ...

[0127] Process 250 can be started at a given frequency, such as once every 20 minutes, once every 10 minutes, once every 5 minutes, or once every 1 minute.

[0128] The method disclosed herein first applies a temperature difference ΔT to a predetermined target, and then applies a temperature T to the internal temperature T of the rack-mounted processing assembly. chip Control is implemented to maintain the liquid flow rate below a predetermined limit (achieved by increasing the liquid flow rate when necessary), and then the air-cooled electronic processing element T is... air cooled IT The temperature is controlled to remain below a predetermined limit (achieved by increasing fan speed when necessary), and then the temperature T of the hot airflow is monitored. air-h Control is implemented to keep the temperature below a predetermined limit (achieved by increasing fan speed when necessary), and finally, the cooling difference is adjusted. c Control measures are implemented to keep the liquid flow rate below predetermined limits (achieved by increasing the liquid flow rate when needed and decreasing the fan speed when needed).

[0129] Through process 250, the temperature difference (ΔT) is optimized and maintained, thereby ensuring better efficiency of the dry cooler unit and keeping the component temperature within an acceptable range.

[0130] Advantageously, ALHEX has a pin c ALHEX can also have a fan speed. In this case, for each ALHEX pin... c The subroutine for fan speed is preferably executed in parallel (instead, preferably there is one flow rack per rack, meaning a pin for checking whether the flow should be increased). c It can be the largest pin c ).

[0131] Now refer to Figures 13 to 15 This describes the technology.

[0132] According to this disclosure, maintenance method 300 is initiated when needed. Advantageously, during the first week of the rack's lifespan, method 300 is initiated daily at the same time (e.g., 10:00 AM), and thereafter once every 30 days at the same time (e.g., 10:00 AM). Obviously, the invention is not limited to this frequency; the operator can determine the frequency based on different parameters (e.g., rack lifespan, rack density in the data center, data center environment, etc.), or method 300 can be initiated when an alarm indicating suspected traffic anomalies is issued. Furthermore, method 300 can be initiated at a time calculated through a prediction step, which will be explained later.

[0133] from Figures 13 to 15 It can be seen that when method 300 starts, operation mode 250 stops at steps 13-1, 14-1, and 15-1.

[0134] Then, maintenance method 300 includes at least the following steps (13-v, 15-v): at a predetermined time (denoted as t) valve Within a predetermined time (T), the intelligent control valve is opened to a fully open state or a predetermined percentage of valve opening; and / or steps (14-f, 15-f): within a predetermined time (denoted as T) fan Within a predetermined time T, at least one fan is operated at maximum speed. valve The time is selected to be between 5 minutes and 30 minutes, preferably, the predetermined time T. valve Selected as 15 minutes. Scheduled time T fan The time is selected to be between 5 minutes and 30 minutes, preferably, the predetermined time T. fan It was selected as 15 minutes.

[0135] exist Figure 13 In the maintenance method 300, only step 13-v of opening the smart valve is included; Figure 14 In the maintenance method 300, only step 14-f of running the fan is included; Figure 15 In the method 300, step 15-v is to open the smart valve and step 15-f is to run the fan.

[0136] As from Figure 14 It can be seen that before step 14-f of running the fan, there is step 14-2, that is, at a predetermined time (denoted as T). stop The fan will stop inside. stop The time is selected to be between 1 second and 10 seconds, preferably, T stop The time was set to 5 seconds.

[0137] Maintenance steps 13-v, 14-f, 15-v, 15-f, and 14-2 ensure automatic cleaning of the liquid cooling pipes of the liquid cooling circuit 105 (by means of maximum liquid flow) and automatic cleaning of the fins of the heat exchanger ALHEX.

[0138] Then, after the maintenance steps are completed, maintenance method 300 includes steps 13-3, 14-3, and 15-3 to restart operation process 250. Figures 13 to 15 This is referred to as "returning to normal state". Finally, exit method 300.

[0139] As can be seen from the figure, method 300 includes steps (warnings) of sending notifications and / or alarms after, before, or during maintenance steps.

[0140] Preferably, the maintenance method 300 includes the step of measuring at least one parameter P of a set of S before, during, and / or after maintenance steps 13-v, 14-f, 15-v, 15-f, 14-2. This set of parameters can be used to assess the condition of the cooling system before, during, and / or after maintenance steps 13-v, 14-f, 15-v, 15-f, 14-2. Specifically, this set of parameters can be used to assess the impact of maintenance steps 13-v, 14-f, 15-v, 15-f, 14-2 on the cooling system 100.

[0141] Advantageously, the parameter P is in ΔT, T R-i T R-o T air-c T air-h Pin c T chip T air cooled IT The choice is between flow rate V and flow rate S. The set of S can be a combination of the parameters P listed above.

[0142] For example, if each, some, or all of the listed parameters decrease during the maintenance steps, the maintenance method can be evaluated as effective. Conversely, if each, some, or all of the listed parameters remain unchanged or increase, the maintenance method can be evaluated as ineffective. Or alternatively, with ΔT and / or T R-o With the increase in [something], this maintenance method can be evaluated as effective.

[0143] For example, the method can compare the difference between the values ​​before and after the maintenance step with one or more predetermined thresholds, where one of the predetermined thresholds may be 0 or not. Depending on the value of the threshold, method 300 may include an efficiency level.

[0144] For example, for parameter P, which should increase with maintenance steps, if the difference D is between 0 and a first (positive) threshold, method 300 can be evaluated as correct; if the difference D is between the first threshold and a second threshold (greater than the first threshold), method 300 can be evaluated as good; and if the difference D is negative, method 300 can be evaluated as inefficient.

[0145] If method 300 is evaluated as invalid, an alert can be sent to the operator.

[0146] For example, for parameter P, which should decrease with maintenance steps, if the difference D is between 0 and a first (negative) threshold, method 300 can be evaluated as correct; if the difference D is between the first threshold and a second threshold (less than the first threshold), method 300 can be evaluated as good; and if the difference D is positive, method 300 can be evaluated as inefficient.

[0147] If method 300 is evaluated as invalid, an alert can be sent to the operator.

[0148] Method 300 preferably includes the step of recording all measured values ​​of the set of S-parameters P. The automatic and systematic recording of the measured values ​​can serve as a basis for predicting the initiation of the next maintenance method. In other words, historical data about maintenance activities is stored.

[0149] For example, after starting maintenance method 300 a certain number of times (e.g., ten times), it can be concluded that during normal use, if each, some, or all of the measured parameters P used in operation method 250 reach a predetermined value, then maintenance method 300 must be started.

[0150] Maintenance method 300 is an automated method that ensures efficient cleaning of the components of the cooling system 100, thereby preventing heat buildup around electronic components, reducing the risk of thermal stress, and keeping the rack clean and well-ventilated at all times, thus ensuring fire safety by reducing the risk of overheating and preventing bacterial growth. Furthermore, method 300 requires no operator intervention or any auxiliary modules or equipment and is cost-effective.

[0151] Furthermore, historical data on maintenance activities can be analyzed to identify trends, optimize maintenance plans, and make informed decisions for future planning. Additionally, Method 300 allows administrators to monitor and manage maintenance activities from a central location, thereby improving efficiency and reducing the need for physical presence in the data center.

[0152] Maintenance method 300 can be used Figure 16The controller 600, as depicted in the high-level functional block diagram, performs the following actions. As shown, the controller 600 includes a processor or multiple cooperating processors (referred to as processor 610 for simplicity), one or more storage devices (referred to as storage device 630 for simplicity), and an input / output interface 620 (or separate input and output interfaces) that allows the controller 600 to communicate with certain components of the liquid-cooled device 100. The processor 610 is operatively connected to the storage device 630 and the input / output interface 620. The storage device 630 includes memory for storing parameters 634, including, for example, but not limited to, the predetermined conductivity threshold described above. The storage device 630 may include a non-transitory computer-readable medium for storing code instructions 632 that can be executed by the processor 610 to allow the controller 600 to perform various tasks assigned to the controller 600.

[0153] The controller 600 is operatively connected to components of the liquid-cooled device 100, such as a temperature sensor for measuring parameter P, via an input / output interface 620. The controller 600 executes code instructions 632 stored in the storage device 630 to implement the aforementioned steps of the maintenance method 300.

[0154] Although the above implementation has been described and illustrated with reference to specific steps performed in a particular order, it will be understood that these steps can be combined, subdivided, or reordered without departing from the teachings of this art. At least some of the steps can be performed in parallel or sequentially. Therefore, the order and grouping of steps are not limitations of this art.

[0155] Modifications and improvements to the above-described implementations of this 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 this technology is intended to be limited only by the scope of the appended claims.

Claims

1. A method for maintaining a system including rack-mounted processing components, the system comprising: Cooling unit (170), the cooling unit being configured to supply cooling fluid to the rack-mounted processing assembly and to receive heated fluid from the rack-mounted processing assembly; A liquid distribution circuit (105) configured to deliver cooling liquid from a dry cooling unit to the rack-mounted processing assembly, the liquid distribution circuit (105) including at least one heat exchanger (ALHEX) configured to use the cooling liquid to cool the airflow of the rack. At least one fan, wherein the at least one fan is used to cool the airflow in the rack; Each of the rack-mounted data processing components includes: At least one heat-generating electronic processing element and at least one liquid cooling block, the at least one liquid cooling block being arranged in thermal contact with a corresponding at least one heat-generating electronic processing element, the at least one liquid cooling block being fluidly coupled to the liquid distribution circuit to receive the cooling liquid and circulate the cooling liquid through the liquid cooling block, and A smart control valve is correspondingly arranged to be fluidly coupled to at least one liquid cooling block of the respective rack-mounted data processing component. The smart control valve is configured to control the flow rate of the cooling fluid of the respective rack-mounted data processing component independently of pressure and based on detected temperature and pressure flow. The flow rate includes a range from a minimum flow rate when the valve is in a state referred to as the closed state to a maximum flow rate when the valve is in a state referred to as the fully open state. The method (250) includes a set of steps referred to as maintenance steps: The step of opening the intelligent control valve to the fully open state during a predetermined time period, and / or The step of running the at least one fan at maximum speed during a predetermined time period; The method further includes: a step of measuring at least one set of parameters before the maintenance step, the set of measured values ​​before the maintenance step being referred to as the initial maintenance set; and a step of measuring the at least one set of parameters after the maintenance step, the set of measured values ​​after the maintenance step being referred to as the final maintenance set.

2. The method according to claim 1, wherein, The predetermined time in the step of opening the intelligent control valve to the fully open state is selected to be between 5 minutes and 30 minutes, preferably 15 minutes.

3. The method according to claim 1 or 2, wherein, The predetermined time in the step of running the at least one fan at maximum speed for a predetermined period of time is selected to be between 5 minutes and 30 minutes, preferably 15 minutes.

4. The method according to any one of the preceding claims, the method comprising: The step of stopping the fan before the step of running the at least one fan at maximum speed.

5. The method according to any one of the preceding claims, wherein, In normal use, the system is managed by a method referred to as the normal use method, which includes the following normal use steps: For the corresponding rack-mounted data processing unit, the current liquid flow rate and the current input coolant temperature (T) are... R-i ) and current output variable heat liquid temperature (T) R-o The measurement is performed, and the current temperature difference (ΔT) between the current input cooling liquid temperature and the current output heat transfer fluid temperature is calculated. The current temperature difference is compared with the target temperature value (254, 256), and the liquid flow rate of the intelligent control valve is dynamically adjusted based on the comparison result, the current liquid flow rate, and the current input cooling liquid temperature. The internal temperature (T) of the rack-mounted processing assembly chip The measurement is performed, and, if the current temperature difference equals the target temperature value, Determine the current input coolant temperature (T) R-i The internal temperature (T) of the rack-mounted processing assembly described below chip Whether it is less than the predetermined limit value (276), and For the internal temperature (T) of at least one air-cooled electronic processing element aircooledIT Measurements were taken of the internal temperature (T) of the rack-mounted processing assembly. chip If the value is less than the predetermined limit (276), and Determine the current input coolant temperature (T) R-i Under these conditions, the internal temperature (T) of the at least one air-cooled electronic processing element aircooledIT Is it less than the predetermined limit value (280)? The maintenance method includes the following steps: before opening the intelligent control valve to the fully open state, at least stop the step of dynamically adjusting the liquid flow rate of the intelligent control valve based on the comparison result, the current liquid flow rate, and the current input cooling liquid temperature; and after the step of opening the intelligent control valve to the fully open state is completed, perform the step of dynamically adjusting the liquid flow rate of the intelligent control valve based on the comparison result, the current liquid flow rate, and the current input cooling liquid temperature.

6. The method according to any one of the preceding claims, the method comprising the steps of sending a notification and / or an alarm after the maintenance step, before the maintenance step, or simultaneously with the maintenance step.

7. The method of claim 1, wherein the method includes the step of recording each measured value of the parameter.

8. The method according to claim 1 or 7, wherein, The method includes evaluating the maintenance steps based on the final maintenance group or based on the difference between the final maintenance group and the initial maintenance group.

9. The method according to any one of claims 1, 7 to 8, wherein, The at least one parameter is selected as the air temperature or the temperature of the heating element at the outlet of the rack, or the input cooling liquid temperature (T). R-i ), and the corresponding rack-mounted data processing component output variable heat liquid temperature (T) R-o The internal temperature (T) of the rack-mounted processing assembly chip ), or the temperature of air-cooled electronic processing components (T) aircooledIT The air temperature downstream of the heat exchanger (T) air-c The air temperature upstream of the heat exchanger (T) air-h ).

10. A computer program comprising instructions that, when executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 9.

11. A computer-readable medium comprising instructions that, when executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 9.

Images