Liquid cooling fault positioning method and device of server, and electronic equipment

Abstract

The invention provides a liquid cooling fault positioning method and device of a server, electronic equipment and a readable storage medium. The liquid cooling fault positioning method of the server comprises the following steps: acquiring real-time temperatures, real-time loads and maximum working temperatures of a plurality of target parts in a target server; according to the real-time temperature, the real-time load and the maximum working temperature, using a preset data model to obtain a cooling liquid flow velocity value of the liquid cooling equipment corresponding to each target component; obtaining a first cooling liquid flow velocity state of each liquid cooling device according to the cooling liquid flow velocity value, a preset maximum flow velocity value and a preset minimum flow velocity value; and in response to determining that an abnormal flow velocity state exists in the first cooling liquid flow velocity states, obtaining a liquid cooling fault positioning result of the target server according to the plurality of first cooling liquid flow velocity states. According to the invention, the purpose of positioning the liquid cooling fault can be achieved, and the accuracy of the obtained liquid cooling fault positioning result can be improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and provides a method, apparatus, electronic device and readable storage medium for locating liquid-cooled faults in servers. Background Technology

[0002] As the computing power of key components such as GPUs and CPUs increases, their power consumption also increases. Servers that deploy these components typically use liquid cooling to dissipate heat, thereby reducing power consumption and improving performance.

[0003] Current technologies typically only identify liquid cooling equipment failures when the component temperature reaches its maximum operating temperature, but they cannot pinpoint the cause of the failure. Therefore, locating liquid cooling failures in servers is a pressing technical problem that needs to be solved. Summary of the Invention

[0004] According to a first aspect of this disclosure, a method for locating liquid cooling faults in a server is provided, comprising: acquiring real-time temperature, real-time load, and maximum operating temperature of multiple target components in a target server; obtaining, using a preset data model, coolant flow rate values ​​of liquid cooling devices corresponding to each target component based on the real-time temperature, the real-time load, and the maximum operating temperature; obtaining a first coolant flow rate state of each liquid cooling device based on the coolant flow rate values, a preset maximum flow rate value, and a preset minimum flow rate value; and, in response to determining that an abnormal flow rate state exists in the first coolant flow rate state, obtaining a liquid cooling fault location result of the target server based on multiple first coolant flow rate states.

[0005] According to a second aspect of this disclosure, a liquid cooling fault location device for a server is provided, comprising: an acquisition unit for acquiring real-time temperature, real-time load, and maximum operating temperature of multiple target components in a target server; an analysis unit for obtaining, based on the real-time temperature, the real-time load, and the maximum operating temperature, a preset data model to obtain coolant flow rate values ​​for liquid cooling devices corresponding to each target component; a processing unit for obtaining a first coolant flow rate state for each liquid cooling device based on the coolant flow rate values, a preset maximum flow rate value, and a preset minimum flow rate value; and a location unit for obtaining a liquid cooling fault location result for the target server based on multiple first coolant flow rate states in response to determining that an abnormal flow rate state exists in the first coolant flow rate state.

[0006] According to a third aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method as described above.

[0007] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform the method described above.

[0008] According to a fifth aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described above.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0011] Figure 1 This is a schematic diagram based on the first embodiment of the present disclosure;

[0012] Figure 2 This is a schematic diagram according to the second embodiment of the present disclosure;

[0013] Figure 3 This is a schematic diagram according to the third embodiment of the present disclosure;

[0014] Figure 4 This is a schematic diagram according to the fourth embodiment of the present disclosure;

[0015] Figure 5 This is a schematic diagram according to the fifth embodiment of the present disclosure;

[0016] Figure 6 This is a block diagram of an electronic device used to implement the liquid cooling fault location method for a server according to embodiments of the present disclosure. Detailed Implementation

[0017] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and mechanisms are omitted in the following description.

[0018] Figure 1 This is a schematic diagram based on the first embodiment of this disclosure. (See diagram below.) Figure 1 As shown, the liquid cooling fault location method for the server in this embodiment specifically includes the following steps:

[0019] S101. Obtain the real-time temperature, real-time load, and maximum operating temperature of multiple target components in the target server;

[0020] S102. Based on the real-time temperature, the real-time load and the maximum operating temperature, a preset data model is used to obtain the coolant flow rate value of the liquid cooling equipment corresponding to each target component.

[0021] S103. Based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value, the first coolant flow rate state of each liquid cooling device is obtained;

[0022] S104. In response to determining that there is an abnormal flow rate state in the first coolant flow rate state, obtain the liquid cooling fault location result of the target server based on multiple first coolant flow rate states.

[0023] The liquid cooling fault location method for servers in this embodiment uses a preset data model to obtain the coolant flow rate value of the liquid cooling device corresponding to the target component based on the real-time temperature, real-time load, and maximum operating temperature of the target component. This achieves the purpose of obtaining the coolant flow rate value of the corresponding liquid cooling device based on the heat dissipation parameters of the component, without the need to add an additional independent component to monitor the liquid cooling device. This improves the accuracy of obtaining the coolant flow rate value, reduces the cost of obtaining the coolant flow rate value, and thus improves the accuracy of the obtained first coolant flow rate state and the accuracy of the liquid cooling fault location result obtained based on the first coolant flow rate state.

[0024] In this embodiment, the server is located in a rack, and a rack may include multiple servers, each of which can be regarded as a node; wherein, the server includes various components, such as CPU, GPU, memory, network card, SSD (Solid State Drive), etc.

[0025] In this embodiment, liquid cooling equipment is used to dissipate heat from the server or its components. Different components in the server can be equipped with different liquid cooling equipment or the same liquid cooling equipment. That is, one liquid cooling equipment can be used to cool the server or its components, or multiple liquid cooling equipment can be used to cool the server or its components.

[0026] In this embodiment, the liquid cooling device can be a cold plate liquid cooling device. One cold plate liquid cooling device can perform liquid cooling on one target component or multiple target components. In this embodiment, the cold plate liquid cooling device is attached to the surface of the target component. By circulating the coolant inside the cold plate, the heat generated by the target component is indirectly removed, thereby achieving the purpose of heat dissipation for the target component.

[0027] In this embodiment, when executing S101, the real-time temperature and real-time load of multiple target components can be obtained through the node monitoring device BMC (Baseboard Management Controller) deployed in the target server.

[0028] In this embodiment, the multiple target components can be all the components included in the target server, or they can be key components in the target server, such as the CPU, GPU, or CPU and GPU in the target server.

[0029] In this embodiment, when executing S101 to obtain the real-time temperature and real-time load of the target component, the component temperature obtained at the current moment can be used as the real-time temperature of the target component, and the component load obtained at the current moment can be used as the real-time load of the target component. In this embodiment, during the operation of the target server, the component temperature and component load corresponding to each target component can be obtained periodically, thereby triggering the subsequent fault location process.

[0030] The maximum operating temperature obtained by executing S101 in this embodiment is a set parameter corresponding to the target component. The maximum operating temperature usually does not change during the operation of the target component. When executing S101 in this embodiment, the maximum operating temperature of the target component can be obtained based on the component name or component type and other identification information of the target component.

[0031] When executing S101, this embodiment may further include the following: in response to determining that the real-time load of the target component has not been obtained, the maximum workload of the corresponding target component is obtained as the real-time load of the target component; wherein, the maximum workload in this embodiment is also a set parameter corresponding to the target component, and the maximum workload of the target component can be obtained according to the component name or component type of the target component.

[0032] In other words, in the absence of real-time load on the target component, this embodiment obtains the maximum working load of the corresponding target component as the real-time load, thereby ensuring that the subsequent fault location process can continue to be executed and avoiding the problem of not being able to obtain the coolant flow rate value due to the lack of real-time load.

[0033] In this embodiment, when executing S101, in addition to obtaining the real-time temperature, real-time load and maximum operating temperature of each target component, the coolant type information can also be obtained. The coolant type information is used to reflect what type of coolant the liquid cooling equipment uses to perform liquid cooling of the target component; that is, under the same conditions, different coolant flow rate values ​​can be obtained according to different coolant type information.

[0034] In this embodiment, after executing S101 to obtain the real-time temperature, real-time load, and maximum operating temperature of multiple target components in the target server, S102 is executed to obtain the coolant flow rate value of the liquid cooling device corresponding to each target component based on the obtained real-time temperature, real-time load, and maximum operating temperature using a preset data model.

[0035] In other words, this embodiment uses a preset data model to obtain the coolant flow rate value of the liquid cooling device corresponding to each target component based on the heat dissipation related parameters (i.e., real-time temperature, real-time load, and maximum operating temperature) of each target component, thereby achieving the purpose of converting the heat dissipation related parameters of the target component into the coolant flow rate value of the liquid cooling device.

[0036] Specifically, in this embodiment, when executing S102 to obtain the coolant flow rate value of the liquid cooling device corresponding to each target component based on the acquired real-time temperature, real-time load, and maximum operating temperature using a preset data model, the implementation method can be as follows: for each target component, a target data model is selected from the preset data models based on the real-time load and maximum operating temperature of the target component, with different real-time loads and different maximum operating temperatures corresponding to different data models; using the selected target data model, the coolant flow rate value of the liquid cooling device corresponding to the target component is obtained based on the real-time temperature of the target component.

[0037] In other words, this embodiment includes different preset data models. The target data model is selected from multiple preset data models based on the real-time load and maximum operating temperature of the target component. This ensures that the selected target data model corresponds to the corresponding load and maximum operating temperature. The target data model is then used to obtain the coolant flow rate value of the liquid cooling equipment corresponding to the target component based on the real-time temperature, thereby improving the accuracy of the obtained coolant flow rate value.

[0038] In this embodiment, the preset data model can be a neural network model. Different preset data models can be trained based on the real-time temperature of the component sample and the sample coolant flow rate. Therefore, after the target data model is selected in S102, the real-time temperature of the target component is input into the target data model. Based on the output of the target data model, the coolant flow rate of the liquid cooling device corresponding to the target component can be obtained.

[0039] In this embodiment, the preset data model can also be a data curve. This data curve is used to reflect the correspondence between the real-time temperature of the component and the coolant flow rate value of the liquid cooling device corresponding to the component. For example, the horizontal axis of the data curve represents the coolant flow rate value, and the vertical axis represents the real-time temperature. Therefore, in this embodiment, when executing S102, the coolant flow rate value corresponding to the real-time temperature in the selected data curve is taken as the coolant flow rate value of the liquid cooling device corresponding to the target component.

[0040] Alternatively, this embodiment may include only a preset data model, which is trained based on the sample real-time temperature, sample maximum operating temperature, sample load, and sample coolant flow rate of the component. That is, during the training process, the sample real-time temperature, sample maximum operating temperature, and sample load are used as inputs to the neural network model. After obtaining the predicted coolant flow rate value output by the neural network model, the parameters of the neural network model are adjusted based on the loss function value calculated from the predicted coolant flow rate value and the sample coolant flow rate value, thereby obtaining the preset data model.

[0041] Therefore, in this embodiment, when executing S102, the real-time temperature, real-time load and maximum operating temperature of each target component can be input into a preset data model, and the output of the preset data model can be used as the coolant flow rate value of the liquid cooling device corresponding to the target component.

[0042] If coolant type information is also obtained when S101 is executed in this embodiment, then when S102 is executed in this embodiment, a target data model can be selected from the preset data model according to the real-time load, maximum operating temperature and coolant type information of the corresponding target component; or the real-time load, real-time temperature, maximum operating temperature and coolant type information of the corresponding target component can be input into the preset data model together to obtain the coolant flow rate value output by the preset data model.

[0043] For example, if the real-time load of the target component is 500W, the real-time temperature is 50℃, and the maximum operating temperature is 65℃, the coolant flow rate value of the liquid cooling device corresponding to the target component obtained by executing S102 in this embodiment can be flow rate value A; if the real-time load of the target component is 500W, the real-time temperature is 60℃, and the maximum operating temperature is 65℃, the coolant flow rate value of the liquid cooling device corresponding to the target component obtained by executing S102 in this embodiment can be flow rate value D; if the real-time load of the target component is 300W, the real-time temperature is 50℃, and the maximum operating temperature is 65℃, the coolant flow rate value of the liquid cooling device corresponding to the target component obtained by executing S102 in this embodiment can be flow rate value C.

[0044] In this embodiment, after executing S102 to obtain the coolant flow rate value of the liquid cooling device corresponding to each target component, S103 is executed to obtain the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value.

[0045] In this embodiment, the first coolant flow rate state of the liquid cooling device includes two types: normal flow rate state and abnormal flow rate state.

[0046] In this embodiment, when executing S103 to obtain the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value, the implementation method can be as follows: obtain the target flow rate value range based on the preset maximum flow rate value and the preset minimum flow rate value; for each liquid cooling device, in response to determining that the coolant flow rate value of the liquid cooling device is within the target flow rate value range, take the normal flow rate state as the first coolant flow rate state of the liquid cooling device; otherwise, take the abnormal flow rate state as the first coolant flow rate state of the liquid cooling device.

[0047] In addition, when executing S103 in this embodiment, the degree of abnormality of the abnormal flow rate state can be obtained based on the flow rate difference between the preset minimum flow rate value and the coolant flow rate value. The larger the flow rate difference, the more severe the abnormality.

[0048] In this embodiment, after executing S103 to obtain the first coolant flow rate status of each liquid cooling device, S104 is executed in response to determining that there is an abnormal flow rate status in the first coolant flow rate status. Based on multiple first coolant flow rate statuses, the liquid cooling fault location result of the target server is obtained.

[0049] In other words, in this embodiment, after determining that there is a faulty liquid cooling device in the target server (i.e., a liquid cooling device with an abnormal flow rate state), the liquid cooling fault cause of the target server is further analyzed based on the multiple first coolant flow rate states corresponding to multiple liquid cooling devices, and then the analyzed liquid cooling fault cause is used as the liquid cooling fault location result.

[0050] In this embodiment, when executing S104, a first flow rate state set can be obtained first based on multiple first coolant flow rate states, and then the liquid cooling fault cause corresponding to the first flow rate state set can be used as the liquid cooling fault location result of the target server.

[0051] In this embodiment, when executing S104, the cause of liquid cooling failure corresponding to the first flow rate state set can be determined according to the first fault table; the first fault table includes different first flow rate state sets and the causes of liquid cooling failure corresponding to different first flow rate state sets.

[0052] For example, if the target component includes component A and component B, if the first coolant flow rate state of the liquid cooling device corresponding to component A is an abnormal flow rate state, and the first coolant flow rate state of the liquid cooling device corresponding to component B is a normal flow rate state, and if the liquid cooling fault cause in the first flow rate state set corresponding to (abnormal, normal) in the first fault table is "poor heat dissipation paste on the cold plate or a problem with the cold plate unit of the corresponding component, rather than a blockage of the cold plate", then in this embodiment, when executing S104, the liquid cooling fault cause is taken as the liquid cooling fault location result of the target server.

[0053] For another example, if the first coolant flow rate state of the liquid cooling device corresponding to component A is an abnormal flow rate state, and the first coolant flow rate state of the liquid cooling device corresponding to component B is an abnormal flow rate state, and if the liquid cooling fault cause of the first flow rate state set corresponding to (abnormal, abnormal) in the first fault table is "cold plate blockage problem", then in this embodiment, when executing S104, the liquid cooling fault cause is used as the liquid cooling fault location result of the target server.

[0054] For another example, if the target server only includes component A, if the first coolant flow rate state of the liquid cooling device corresponding to component A is an abnormal flow rate state, and if the liquid cooling fault cause of the first flow rate state set corresponding to (abnormal, empty) in the first fault table is "cold plate blockage or poor cold plate heat dissipation paste", then in this embodiment, when executing S104, the liquid cooling fault cause is taken as the liquid cooling fault location result of the target server.

[0055] Therefore, this embodiment achieves the purpose of accurately locating the faults of the liquid cooling equipment in the target server based on the obtained multiple first coolant flow rate states. It can solve the problem that the existing technology cannot determine the cause of liquid cooling faults when liquid cooling equipment faults occur, thereby improving the efficiency of liquid cooling fault resolution and enhancing the operating performance of the server.

[0056] In addition, when executing S104, this embodiment may also include the following: determining the target component corresponding to the liquid cooling device in an abnormal flow rate state; generating early warning information corresponding to the determined target component for early warning.

[0057] In other words, this embodiment will issue an early warning even if the real-time temperature of the target component corresponding to the liquid cooling equipment has not reached the maximum operating temperature when it is determined that the liquid cooling equipment is in an abnormal flow rate state. This provides a longer maintenance time window so that business scheduling can be carried out in a timely manner, avoiding direct component downtime and affecting business.

[0058] Figure 2 This is a schematic diagram according to the second embodiment of this disclosure. (See diagram below.) Figure 2 As shown, the liquid cooling fault location method for the server in this embodiment may further include the following:

[0059] S201. Determine the target server rack corresponding to the target server;

[0060] S202. Obtain the second coolant flow rate status of multiple first servers in the target cabinet;

[0061] S203. Based on the second coolant flow rate status of the plurality of first servers, obtain the liquid cooling fault location result of the target cabinet.

[0062] In other words, after completing the liquid cooling fault location of the target server, this embodiment can further obtain the liquid cooling fault location result of the target cabinet where the target server is located, thereby improving the comprehensiveness of the obtained liquid cooling fault location result and achieving the purpose of accurately locating the liquid cooling fault of the target cabinet.

[0063] In this embodiment, a rack includes multiple servers, and a server can be regarded as a node, that is, the rack includes multiple nodes.

[0064] In this embodiment, the target cabinet determined in S201 is the cabinet where the target server is located.

[0065] In this embodiment, when executing S202 to obtain the second coolant flow rate status of multiple first servers in the target rack, the implementation method can be as follows: for each first server, obtain multiple first coolant flow rate statuses corresponding to the first server, that is, the first coolant flow rate status corresponds to the liquid cooling device in the first server; in response to determining that multiple first coolant flow rate statuses are all abnormal flow rate statuses, determine that the second coolant flow rate status of the first server is an abnormal flow rate status, otherwise determine that the second coolant flow rate status of the first server is a normal flow rate status, that is, the second coolant flow rate status corresponds to the first server.

[0066] In other words, this embodiment obtains the second coolant flow rate state corresponding to the first server based on the first coolant flow rate state of the liquid cooling equipment in the first server, thereby achieving the purpose of obtaining the coolant flow rate state of the corresponding server based on the coolant flow rate state of the liquid cooling equipment in the server. This can improve the accuracy of the obtained second coolant flow rate state, and thus achieve the purpose of obtaining the liquid cooling fault location result of the target cabinet based on the second coolant flow rate state.

[0067] In this embodiment, when executing S203, a second flow rate state set can be obtained first based on multiple second coolant flow rate states, and then the liquid cooling fault cause corresponding to the second flow rate state set can be used as the liquid cooling fault location result of the target cabinet.

[0068] In this embodiment, when S203 is executed, the cause of liquid cooling failure corresponding to the second flow rate state set can be determined according to the second fault table; the second fault table includes different second flow rate state sets and the causes of liquid cooling failure corresponding to different second flow rate state sets.

[0069] For example, if the target cabinet includes a first server 1 and a first server 2, if the second coolant flow rate status of the first server 1 is an abnormal flow rate status and the second coolant flow rate status of the first server 2 is a normal flow rate status, and if the liquid cooling fault cause corresponding to the (abnormal, normal) second flow rate status set in the second fault table is "single server cold plate fault", then in this embodiment, when executing S203, the liquid cooling fault cause is taken as the liquid cooling fault location result of the target cabinet.

[0070] For another example, if the second coolant flow rate status of the first server 1 is an abnormal flow rate status, and the second coolant flow rate status of the first server 2 is an abnormal flow rate status, and if the liquid cooling fault cause of the corresponding (abnormal, abnormal) second flow rate status set in the second fault table is "single cabinet liquid cooling pipe fault (possible cabinet water distributor fault)", then in this embodiment, when executing S203, the liquid cooling fault cause is taken as the liquid cooling fault location result of the target cabinet.

[0071] After executing S203, this embodiment may further include the following: determining the non-target rack corresponding to the target server, where the non-target rack is another rack in the data center that is in the same column or a different column from the target rack; obtaining the second coolant flow rate status of the second server in the non-target rack, where the second server is the server located in the non-target rack; in response to determining that the second coolant flow rate status of both the first server and the second server is an abnormal flow rate status, obtaining a cooling distribution unit (CDU) fault or a link pipe fault as the liquid cooling fault location result of the target data center; in this embodiment, the target data center is the data center where the target server is located.

[0072] In other words, this embodiment can not only locate liquid cooling faults in a single cabinet, but also in multiple cabinets, that is, locate liquid cooling faults in the computer room, thereby further improving the comprehensiveness of fault location.

[0073] Figure 3 This is a schematic diagram according to the third embodiment of the present disclosure. Figure 3 The diagram shows the architecture of the cold plate liquid cooling system for the data center in this embodiment; Figure 3 In the data center, the server room includes two liquid-cooled cabinets, liquid-cooled cabinet 1 and liquid-cooled cabinet 2. Each liquid-cooled cabinet contains N nodes (each node is a server), and each node is liquid-cooled using cold plate liquid cooling equipment.

[0074] If the target server is a node in liquid-cooled rack 1, this embodiment can first locate the liquid cooling failure of the target server, then locate the liquid cooling failure of liquid-cooled rack 1 based on the first server included in liquid-cooled rack 1 (i.e., the node in liquid-cooled rack 1), and finally locate the liquid cooling failure of the data center where the liquid-cooled rack is located based on the first server in liquid-cooled rack 1 and the second server in liquid-cooled rack 2 (i.e., the node in liquid-cooled rack 2).

[0075] In other words, this embodiment can achieve the purpose of liquid cooling fault location at different granularities for servers, server racks, and server rooms, thereby improving the comprehensiveness of liquid cooling fault location.

[0076] Figure 4 This is a schematic diagram according to the fourth embodiment of the present disclosure. Figure 4The diagram shows a schematic of the preset data model of this embodiment. In this embodiment, the preset data model consists of different data curves, such as a data curve corresponding to a real-time load of 500W, a data curve corresponding to a real-time load of 450W, etc. Different data curves reflect the relationship between the real-time temperature of the component and the coolant flow rate of the liquid cooling equipment corresponding to the component under different real-time loads (this embodiment is described with the maximum operating temperature being the same). In this embodiment, the flow rate range between the preset maximum flow rate value and the preset minimum flow rate value is the target flow rate range when the corresponding component is working normally.

[0077] Therefore, in this embodiment, when determining the coolant flow rate value of the corresponding liquid cooling equipment according to the preset data model, the corresponding data curve is first selected based on the component load and the maximum operating temperature of the component. Then, the coolant flow rate value corresponding to the component temperature in the selected data curve is used as the coolant flow rate value of the liquid cooling equipment corresponding to the component, thereby improving the accuracy of the obtained coolant flow rate value and further improving the accuracy of the fault location result obtained based on the coolant flow rate value.

[0078] Figure 5 This is a schematic diagram according to the fifth embodiment of this disclosure. (See diagram below.) Figure 5 As shown, the liquid cooling fault location device 500 for the server in this embodiment includes:

[0079] The acquisition unit 501 is used to acquire the real-time temperature, real-time load, and maximum operating temperature of multiple target components in the target server;

[0080] Analysis unit 502 is used to obtain the coolant flow rate value of the liquid cooling equipment corresponding to each target component based on the real-time temperature, the real-time load and the maximum operating temperature using a preset data model;

[0081] Processing unit 502 is used to obtain the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value and the preset minimum flow rate value;

[0082] The positioning unit 503 is used to respond to the determination that there is an abnormal flow rate state in the first coolant flow rate state, and to obtain the liquid cooling fault location result of the target server based on multiple first coolant flow rate states.

[0083] The acquisition unit 501 can acquire the real-time temperature and real-time load of multiple target components through the node monitoring device BMC (Baseboard Management Controller) deployed in the target server.

[0084] In this embodiment, the multiple target components can be all the components included in the target server, or they can be key components in the target server, such as the CPU, GPU, or CPU and GPU in the target server.

[0085] In this embodiment, when acquiring the real-time temperature and real-time load of the target component, the acquisition unit 501 can take the component temperature acquired at the current moment as the real-time temperature of the target component and the component load acquired at the current moment as the real-time load of the target component. In this embodiment, during the operation of the target server, the component temperature and component load corresponding to the target component can be acquired periodically for each target component, thereby triggering the subsequent fault location process.

[0086] The maximum operating temperature acquired by the acquisition unit 501 is a set parameter corresponding to the target component. The maximum operating temperature usually does not change during the operation of the target component. The acquisition unit 501 can acquire the maximum operating temperature of the target component based on the component name or component type and other identification information of the target component.

[0087] The acquisition unit 501 may further include the following: in response to determining that the real-time load of the target component has not been acquired, the maximum workload of the corresponding target component is acquired as the real-time load of the target component; wherein, the maximum workload in this embodiment is also a set parameter corresponding to the target component, and the maximum workload of the target component can be acquired according to the component name or component type of the target component.

[0088] In other words, if the acquisition unit 501 is unable to acquire the real-time load of the target component, it acquires the maximum working load of the corresponding target component as the real-time load, thereby ensuring that the subsequent fault location process can continue to be executed and avoiding the problem of not being able to obtain the coolant flow rate value due to the failure to acquire the real-time load.

[0089] In addition to acquiring the real-time temperature, real-time load, and maximum operating temperature of each target component, the acquisition unit 501 can also acquire coolant type information. The coolant type information is used to reflect what type of coolant the liquid cooling equipment uses to perform liquid cooling on the target component.

[0090] In this embodiment, after the acquisition unit 501 acquires the real-time temperature, real-time load, and maximum operating temperature of multiple target components in the target server, the analysis unit 502 uses a preset data model based on the acquired real-time temperature, real-time load, and maximum operating temperature to obtain the coolant flow rate value of the liquid cooling device corresponding to each target component.

[0091] In other words, the analysis unit 502 uses a preset data model to obtain the coolant flow rate value of the liquid cooling device corresponding to each target component based on the heat dissipation related parameters (i.e., real-time temperature, real-time load, and maximum operating temperature) of each target component, thereby achieving the purpose of converting the heat dissipation related parameters of the target component into the coolant flow rate value of the liquid cooling device.

[0092] Specifically, when the analysis unit 502 obtains the coolant flow rate value of the liquid cooling equipment corresponding to each target component based on the acquired real-time temperature, real-time load, and maximum operating temperature using a preset data model, the following implementation method can be adopted: for each target component, select a target data model from the preset data models based on the real-time load and maximum operating temperature of the target component; use the selected target data model to obtain the coolant flow rate value of the liquid cooling equipment corresponding to the target component based on the real-time temperature of the target component.

[0093] In other words, this embodiment includes different preset data models. The analysis unit 502 selects a target data model from multiple preset data models based on the real-time load and maximum operating temperature of the target component, so that the selected target data model can correspond to the corresponding load and maximum operating temperature. Then, the target data model is used to obtain the coolant flow rate value of the liquid cooling equipment corresponding to the target component based on the real-time temperature, which can improve the accuracy of the obtained coolant flow rate value.

[0094] In this embodiment, the preset data model can be a neural network model. Different preset data models can be trained based on the real-time temperature of the component sample and the sample coolant flow rate. Therefore, the analysis unit 502 can input the real-time temperature of the target component into the selected target data model, and obtain the coolant flow rate value of the liquid cooling device corresponding to the target component based on the output result of the target data model.

[0095] In this embodiment, the preset data model can also be a data curve, which is used to reflect the correspondence between the real-time temperature of the component and the coolant flow rate of the liquid cooling device corresponding to the component. For example, the horizontal axis of the data curve represents the coolant flow rate and the vertical axis represents the real-time temperature. Therefore, the analysis unit 502 takes the coolant flow rate corresponding to the real-time temperature in the selected data curve as the coolant flow rate of the liquid cooling device corresponding to the target component.

[0096] Alternatively, this embodiment may include only a preset data model, which is trained based on the sample real-time temperature, sample maximum operating temperature, sample load, and sample coolant flow rate of the component. That is, during the training process, the sample real-time temperature, sample maximum operating temperature, and sample load are used as inputs to the neural network model. After obtaining the predicted coolant flow rate value output by the neural network model, the parameters of the neural network model are adjusted based on the loss function value calculated from the predicted coolant flow rate value and the sample coolant flow rate value, thereby obtaining the preset data model.

[0097] Therefore, the analysis unit 502 can also input the real-time temperature, real-time load and maximum operating temperature of each target component into a preset data model, and use the output of the preset data model as the coolant flow rate value of the liquid cooling equipment corresponding to the target component.

[0098] If the acquisition unit 501 also acquires coolant type information, the analysis unit 502 can select a target data model from the preset data model based on the real-time load, maximum operating temperature and coolant type information of the corresponding target component; or it can input the real-time load, real-time temperature, maximum operating temperature and coolant type information of the corresponding target component into the preset data model together to obtain the coolant flow rate value output by the preset data model.

[0099] In this embodiment, after the analysis unit 502 obtains the coolant flow rate value of the liquid cooling device corresponding to each target component, the processing unit 503 obtains the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value.

[0100] In this embodiment, the first coolant flow rate state of the liquid cooling device includes two types: normal flow rate state and abnormal flow rate state.

[0101] When processing unit 503 obtains the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value, the following implementation method can be adopted: obtain the target flow rate value range based on the preset maximum flow rate value and the preset minimum flow rate value; for each liquid cooling device, in response to determining that the coolant flow rate value of the liquid cooling device is within the target flow rate value range, take the normal flow rate state as the first coolant flow rate state of the liquid cooling device; otherwise, take the abnormal flow rate state as the first coolant flow rate state of the liquid cooling device.

[0102] In addition, the processing unit 503 can also determine the degree of abnormality of the abnormal flow rate state based on the flow rate difference between the preset minimum flow rate value and the coolant flow rate value. The larger the flow rate difference, the more severe the abnormality.

[0103] In this embodiment, after the processing unit 503 obtains the first coolant flow rate status of each liquid cooling device, the positioning unit 504 responds to the determination that there is an abnormal flow rate status in the first coolant flow rate status, and obtains the liquid cooling fault location result of the target server based on multiple first coolant flow rate statuses.

[0104] In other words, when the positioning unit 504 determines that there is a faulty liquid cooling device in the target server (i.e., a liquid cooling device with an abnormal flow rate state), it further analyzes the cause of the liquid cooling fault of the target server based on the multiple first coolant flow rate states corresponding to multiple liquid cooling devices, and then uses the analyzed cause of the liquid cooling fault as the liquid cooling fault location result.

[0105] The positioning unit 504 can first obtain a first flow rate state set based on multiple first coolant flow rate states, and then use the liquid cooling fault cause corresponding to the first flow rate state set as the liquid cooling fault positioning result of the target server.

[0106] The positioning unit 504 can determine the cause of liquid cooling failure corresponding to the first flow rate state set based on the first fault table; the first fault table includes different first flow rate state sets and the causes of liquid cooling failure corresponding to different first flow rate state sets.

[0107] Therefore, the positioning unit 504 achieves the purpose of accurately locating the fault of the liquid cooling equipment in the target server based on the obtained multiple first coolant flow rate states. This can solve the problem that the existing technology cannot determine the cause of liquid cooling fault when liquid cooling equipment fails, thereby improving the efficiency of liquid cooling fault resolution and enhancing the operating performance of the server.

[0108] In addition, the positioning unit 504 may also include the following: identifying the target component corresponding to the liquid cooling equipment in an abnormal flow rate state; generating early warning information corresponding to the identified target component for early warning.

[0109] In other words, when the positioning unit 504 determines that the liquid cooling equipment is in an abnormal flow rate state, it will issue an early warning even if the real-time temperature of the target component corresponding to the liquid cooling equipment has not reached the maximum operating temperature. This provides a longer maintenance time window so that business scheduling can be carried out in a timely manner, avoiding direct component downtime and impact on business.

[0110] The positioning unit 504 can also perform the following: determine the target cabinet of the corresponding target server; obtain the second coolant flow rate status of multiple first servers in the target cabinet; and obtain the liquid cooling fault location result of the target cabinet based on the second coolant flow rate status of multiple first servers.

[0111] In other words, after completing the liquid cooling fault location of the target server, the location unit 504 can further obtain the liquid cooling fault location result of the target cabinet where the target server is located, thereby improving the comprehensiveness of the obtained liquid cooling fault location result and achieving the purpose of accurately locating the liquid cooling fault of the target cabinet.

[0112] In this embodiment, a rack includes multiple servers, and a server can be regarded as a node, that is, the rack includes multiple nodes.

[0113] The target cabinet identified by the positioning unit 504 is the cabinet where the target server is located.

[0114] When the positioning unit 504 acquires the second coolant flow rate status of multiple first servers in the target rack, it can adopt the following implementation method: for each first server, acquire multiple first coolant flow rate statuses corresponding to the first server, that is, the first coolant flow rate status corresponds to the liquid cooling device in the first server; in response to determining that multiple first coolant flow rate statuses are all abnormal flow rate statuses, determine that the second coolant flow rate status of the first server is an abnormal flow rate status, otherwise determine that the second coolant flow rate status of the first server is a normal flow rate status, that is, the second coolant flow rate status corresponds to the first server.

[0115] In other words, the positioning unit 504 obtains the second coolant flow rate state corresponding to the first server based on the first coolant flow rate state of the liquid cooling equipment in the first server, thereby achieving the purpose of obtaining the coolant flow rate state of the corresponding server based on the coolant flow rate state of the liquid cooling equipment in the server. This can improve the accuracy of the obtained second coolant flow rate state, and thus achieve the purpose of obtaining the liquid cooling fault location result of the target cabinet based on the second coolant flow rate state.

[0116] The positioning unit 504 can first obtain a second flow rate state set based on multiple second coolant flow rate states, and then use the liquid cooling fault cause corresponding to the second flow rate state set as the liquid cooling fault positioning result of the target cabinet.

[0117] The positioning unit 504 can determine the cause of liquid cooling failure corresponding to the second flow rate state set based on the second fault table; the second fault table includes different second flow rate state sets and the causes of liquid cooling failure corresponding to different second flow rate state sets.

[0118] The positioning unit 504 can also perform the following: determine the non-target cabinet corresponding to the target server, where the non-target cabinet is another cabinet in the computer room that is in the same column or different column from the target cabinet; obtain the second coolant flow rate status of the second server in the non-target cabinet, where the second server is the server located in the non-target cabinet; in response to determining that the second coolant flow rate status of both the first server and the second server is an abnormal flow rate status, obtain a cooling distribution unit (CDU) fault or a link pipe fault as the liquid cooling fault location result of the target computer room; in this embodiment, the target computer room is the computer room where the target server is located.

[0119] In other words, the positioning unit 504 can not only locate liquid cooling faults in a single rack, but also in multiple racks, i.e., locate liquid cooling faults in the entire data center, thereby further improving the comprehensiveness of fault location. The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein all comply with relevant laws and regulations and do not violate public order and good morals.

[0120] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0121] like Figure 6 The diagram shown is a block diagram of an electronic device for a liquid-cooled fault location method for a server according to an embodiment of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0122] like Figure 6 As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded from storage unit 608 into random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0123] Multiple components in device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of displays, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0124] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as a liquid-cooled fault location method for a server. For example, in some embodiments, the liquid-cooled fault location method for a server may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608.

[0125] In some embodiments, part or all of the computer program may be loaded and / or installed on the device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by computing unit 601, one or more steps of the liquid cooling fault location method for the server described above may be performed. Alternatively, in other embodiments, computing unit 601 may be configured to perform the liquid cooling fault location method for the server by any other suitable means (e.g., by means of firmware).

[0126] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0127] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to the processor or controller of a liquid-cooled fault location device of a general-purpose computer, special-purpose computer, or other programmable server, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0128] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0129] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for showing information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0130] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0131] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0132] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0133] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for locating liquid cooling faults in a server, comprising: Obtain the real-time temperature, real-time load, and maximum operating temperature of multiple target components in the target server; Based on the real-time temperature, the real-time load, and the maximum operating temperature, a preset data model is used to obtain the coolant flow rate value of the liquid cooling equipment corresponding to each target component. Based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value, the first coolant flow rate state of each liquid cooling device is obtained; In response to the determination of an abnormal flow rate state in the first coolant flow rate state, the liquid cooling fault location result of the target server is obtained based on multiple first coolant flow rate states.

2. The method according to claim 1, wherein, The step of obtaining the coolant flow rate value of the liquid cooling equipment corresponding to each target component using a preset data model based on the real-time temperature, the real-time load, and the maximum operating temperature includes: For each target component, a target data model is selected from the preset data models based on the real-time load and maximum operating temperature of that target component. Using the target data model, the coolant flow rate of the liquid cooling equipment corresponding to the target component is obtained based on the real-time temperature of the target component.

3. The method according to claim 1, wherein, The step of obtaining the coolant flow rate value of the liquid cooling equipment corresponding to each target component using a preset data model based on the real-time temperature, the real-time load, and the maximum operating temperature includes: For each target component, the real-time temperature, real-time load, and maximum operating temperature of that target component are input into the preset data model; Based on the output of the preset data model, the coolant flow rate value of the liquid cooling device corresponding to the target component is obtained.

4. The method according to claim 1, wherein, The step of obtaining the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value, and the preset minimum flow rate value includes: The target flow velocity range is obtained based on the preset maximum flow velocity value and the preset minimum flow velocity value; For each liquid cooling device, in response to determining that the coolant flow rate value of the liquid cooling device is within the target flow rate value range, the normal flow rate state is taken as the first coolant flow rate state of the liquid cooling device; otherwise, the abnormal flow rate state is taken as the first coolant flow rate state of the liquid cooling device.

5. The method according to claim 1, wherein, The process of obtaining the liquid cooling fault location result of the target server based on multiple first coolant flow rate states includes: Based on the plurality of first coolant flow rate states, a first flow rate state set is obtained; The cause of the liquid cooling failure corresponding to the first flow rate state set is used as the liquid cooling failure location result of the target server.

6. The method according to claim 1, further comprising: Identify the target component corresponding to the liquid cooling equipment in an abnormal flow rate state; Generate early warning information corresponding to the identified target component to provide early warning.

7. The method according to claim 1, further comprising: Determine the target rack corresponding to the target server; Obtain the second coolant flow rate status of multiple first servers in the target rack; Based on the second coolant flow rate status of the plurality of first servers, the liquid cooling fault location result of the target rack is obtained.

8. The method according to claim 7, wherein, The step of obtaining the second coolant flow rate status of multiple first servers in the target rack includes: For each first server, obtain multiple first coolant flow rate states corresponding to that first server; In response to determining that all of the plurality of first coolant flow rate states are abnormal flow rate states, the second coolant flow rate state of the first server is determined to be an abnormal flow rate state; otherwise, the second coolant flow rate state of the first server is determined to be a normal flow rate state.

9. The method of claim 7, further comprising: Identify the non-target rack corresponding to the target server; Obtain the second coolant flow rate status of the second server in the non-target rack; In response to determining that the second coolant flow rate status of both the first server and the second server is abnormal, a fault in the cooling capacity distribution unit or a fault in the link pipe is obtained as the liquid cooling fault location result of the target computer room.

10. The method according to claim 1, further comprising: In response to determining that the real-time load of the target component is not obtained, the maximum workload of the target component is obtained as the real-time load of the target component.

11. A liquid-cooled fault location device for a server, comprising: The acquisition unit is used to acquire the real-time temperature, real-time load, and maximum operating temperature of multiple target components in the target server. The analysis unit is used to obtain the coolant flow rate value of the liquid cooling equipment corresponding to each target component based on the real-time temperature, the real-time load and the maximum operating temperature, using a preset data model. The processing unit is used to obtain the first coolant flow rate state of each liquid cooling device based on the coolant flow rate value, the preset maximum flow rate value and the preset minimum flow rate value; The positioning unit is used to respond to the determination that there is an abnormal flow rate state in the first coolant flow rate state, and to obtain the liquid cooling fault positioning result of the target server based on multiple first coolant flow rate states.

12. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method of any one of claims 1-10.

13. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-10.

14. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-10.

Images