Temperature control prediction method, device and equipment of liquid cooling server and medium

Abstract

The invention relates to a temperature control prediction method, device and equipment for a liquid cooling server and a medium, and belongs to the field of server temperature control. The method comprises the following steps: acquiring a historical monitoring data set of the liquid cooling server; constructing a liquid cooling server temperature prediction model based on the historical monitoring data set of the liquid cooling server through an improved XGBOOST algorithm; acquiring a preset parameter set of the liquid cooling server; inputting a liquid cooling server preset parameter set into the liquid cooling server temperature prediction model to obtain predicted temperature and predicted temperature uniformity; and constructing a temperature control model of the liquid cooling server based on the predicted temperature and the predicted temperature uniformity, outputting temperature control information, and feeding back a parameter adjustment instruction of the liquid cooling server based on the temperature control information. And measures are taken to control the two within a reasonable range.

Description

Technical Field

[0001] The present invention belongs to the technical field of server temperature control, and in particular relates to a temperature control prediction method, device, equipment and medium for a liquid cooling server. Background Art

[0002] As data centers expand, server density continues to increase, and traditional air cooling systems can no longer meet the needs of high-density, high-performance computing. Liquid cooling technology is widely used in high-performance servers due to its efficient heat conduction capabilities. However, liquid cooling systems require precise control of the flow rate and temperature of the liquid to prevent the equipment from overheating. Existing temperature control systems are mostly based on static thresholds and empirical rules, and lack the ability to predict future states. Traditional temperature control systems are often reactive, that is, they only start to adjust after detecting that the temperature exceeds the standard. This method may not be able to prevent temperature anomalies in time; secondly, it is inefficient: the lack of prediction function means that the system cannot actively adjust the cooling strategy according to changes in server load, which may lead to energy waste. In addition, traditional temperature control systems often ignore the uniformity of temperature between servers, which can easily lead to system failures due to overheating of some components.

[0003] XGBoost (gradient boosting algorithm) has the advantages of processing large-scale data sets and solving complex prediction problems. It is an algorithm that can meet a variety of machine learning tasks such as classification, regression, sorting, and custom objective functions. It sorts the data according to the features before training, and uses a sparse matrix storage format for storage in each block structure. The block structure will be used repeatedly in the subsequent training process, which can greatly reduce the amount of calculation. XGBoost not only supports gradient boosting decision trees but also linear classifiers, thus providing options with application value in a wide range of scenarios. The objective function of XGBoost consists of two parts, namely the loss function and the regularization term. The addition of the regularization term can control the complexity of the model, making the model simpler and helping to prevent overfitting. The forward distribution algorithm is used for greedy training. In each iteration, a CART tree is learned to fit the prediction residuals of the first t-1 trees with the actual values ​​of the training samples. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a temperature control prediction method, device, equipment and medium for a liquid cooling server.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A temperature control prediction method for a liquid cooling server, the implementation of the temperature control prediction method comprising the following steps:

[0007] Acquire a historical monitoring data set of a liquid cooling server, wherein the historical monitoring data set of the liquid cooling server includes a historical temperature of the liquid cooling server, a historical operating computing power of the liquid cooling server, a historical operating power of the liquid cooling server, a historical operating primary oil pump speed of the liquid cooling server, a historical operating secondary water pump speed of the liquid cooling server, and a historical operating environment temperature of the liquid cooling server;

[0008] Building a temperature prediction model for a liquid cooling server based on the historical monitoring data set of the liquid cooling server by improving the XGBOOST algorithm;

[0009] Obtaining a preset parameter set of a liquid cooling server, wherein the preset parameter set of the liquid cooling server includes a liquid cooling server operating computing power, a liquid cooling server operating power, a liquid cooling server operating primary oil pump speed, a liquid cooling server operating secondary water pump speed, and a liquid cooling server operating environment temperature;

[0010] Inputting the preset parameter set of the liquid cooling server into the temperature prediction model of the liquid cooling server to obtain the predicted temperature and the predicted temperature uniformity;

[0011] Based on the predicted temperature and the predicted temperature uniformity, a liquid cooling server temperature control model is constructed and temperature control information is output, wherein the temperature control information includes temperature control abnormality information and temperature control normal information. Based on the temperature control information, liquid cooling server parameter adjustment instructions are fed back, wherein the liquid cooling server parameter adjustment instructions include primary oil pump speed adjustment instructions, secondary water pump speed adjustment instructions, and operating computing power adjustment instructions.

[0012] Preferably, the step of constructing a temperature prediction model for a liquid cooling server based on the historical monitoring data set of the liquid cooling server by improving the XGBOOST algorithm includes:

[0013] Acquire the historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server, wherein the historical temperature of the liquid cooling server includes the historical temperature of the CPU, the historical temperature of the GPU, and the historical temperature of the memory;

[0014] Given the historical monitoring data set of the liquid cooling server and the historical temperature uniformity of the liquid cooling server, the improved XGBOOST algorithm is used to perform additive learning of k decision trees to construct the temperature prediction model of the liquid cooling server.

[0015] Preferably, obtaining the historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server includes:

[0016] Obtain the number of heat-generating components in the liquid-cooled server, where the number of heat-generating components includes the number of CPUs, the number of GPUs, and the number of memories;

[0017] The historical temperature uniformity of the liquid cooling server is calculated based on the historical temperature of the liquid cooling server and the number of the heat generating components.

[0018] Preferably, the step of performing additive learning of k decision trees through the improved XGBOOST algorithm to construct the temperature prediction model of the liquid cooling server includes:

[0019] Obtaining a loss function of the temperature prediction model of the liquid cooling server, wherein the loss function includes a temperature loss function and a temperature uniformity loss function;

[0020] Obtaining a model regularization term of the liquid cooling server temperature prediction model;

[0021] Obtaining a model objective function based on the loss function and the model regularization term;

[0022] Expand the model objective function by a second-order Taylor expansion;

[0023] The model objective function is simplified by simplifying constant terms and the liquid cooling server temperature prediction model is obtained.

[0024] Preferably, the expression of the temperature loss function is: Among them, l is the temperature loss function, y j is the historical temperature of the liquid-cooled server, is the predicted historical temperature output by the j-th decision tree of the model at the t-th iteration, is the predicted historical temperature output by the jth decision tree of the model at the t-1th iteration, is the predicted historical temperature of the jth decision tree in the model at the tth iteration, k is the number of decision trees in the model; the expression of the temperature uniformity loss function is Where w is the temperature uniformity loss function, r j For the historical temperature uniformity of liquid-cooled servers, is the predicted historical temperature uniformity output by the j-th decision tree of the model at the t-th iteration, is the predicted historical temperature uniformity output by the jth decision tree of the model at the t-1th iteration, is the predicted historical temperature uniformity of the jth decision tree of the model at the tth iteration.

[0025] Preferably, the step of constructing a temperature control model for a liquid cooling server based on the predicted temperature and the predicted temperature uniformity and outputting temperature control information comprises:

[0026] A predicted temperature threshold and a predicted temperature uniformity threshold are preset. When the predicted temperature is greater than or equal to the predicted temperature threshold or the predicted temperature uniformity is less than or equal to the predicted temperature uniformity threshold, the temperature control abnormality information is output; when the predicted temperature is less than the predicted temperature threshold and the predicted temperature uniformity is greater than the predicted temperature uniformity threshold, the temperature control normal information is output.

[0027] Preferably, the liquid cooling server parameter adjustment instruction based on the temperature control information feedback includes:

[0028] When the temperature control abnormality information is output, the speed of the oil pump of the liquid cooling server is judged, and when the speed of the oil pump of the liquid cooling server is less than the maximum speed of the oil pump, the speed adjustment instruction of the oil pump is fed back to adjust the speed of the oil pump of the liquid cooling server until the normal temperature control information is output;

[0029] When the speed of the primary oil pump of the liquid-cooled server is equal to the maximum speed, the speed of the secondary water pump of the liquid-cooled server is determined; when the speed of the secondary water pump of the liquid-cooled server is less than the maximum speed of the secondary water pump, the speed adjustment instruction of the secondary water pump is fed back to adjust the speed of the secondary water pump of the liquid-cooled server until the normal temperature control information is output;

[0030] When the speed of the secondary water pump of the liquid cooling server is equal to the maximum speed, the operation computing power adjustment instruction is fed back to reduce the operation computing power of the liquid cooling server until the temperature control normal information is output.

[0031] A temperature control prediction device for a liquid cooling server, used to execute the temperature control prediction method described above, comprising a data acquisition module, a model building module, and a temperature control module;

[0032] The data acquisition module is used to acquire a historical monitoring data set of a liquid cooling server, and the historical monitoring data set of the liquid cooling server includes a historical temperature of the liquid cooling server, a historical operating computing power of the liquid cooling server, a historical operating power of the liquid cooling server, a historical operating primary oil pump speed of the liquid cooling server, a historical operating secondary water pump speed of the liquid cooling server, and a historical operating environment temperature of the liquid cooling server;

[0033] The model building module is used to build a liquid cooling server temperature prediction model based on the liquid cooling server historical monitoring data set by improving the XGBOOST algorithm; obtain a liquid cooling server preset parameter set, the liquid cooling server preset parameter set includes the liquid cooling server operation computing power, the liquid cooling server operation power, the liquid cooling server operation primary oil pump speed, the liquid cooling server operation secondary water pump speed, and the liquid cooling server operation environment temperature; the liquid cooling server preset parameter set is input into the liquid cooling server temperature prediction model to obtain the predicted temperature and the predicted temperature uniformity;

[0034] The temperature control module is used to build a liquid cooling server temperature control model based on the predicted temperature and the predicted temperature uniformity and output temperature control information, wherein the temperature control information includes temperature control abnormality information and temperature control normal information, and to feedback liquid cooling server parameter adjustment instructions based on the temperature control information, wherein the liquid cooling server parameter adjustment instructions include primary oil pump speed adjustment instructions, secondary water pump speed adjustment instructions, and operating computing power adjustment instructions.

[0035] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the temperature control prediction method described above is implemented when the processor executes the program.

[0036] A storage medium comprising computer executable instructions, wherein the computer executable instructions are used to execute the temperature control prediction method described above when executed by a computer processor.

[0037] The beneficial effects of the present invention are:

[0038] (1) By improving the XGBOOST algorithm and building a temperature prediction model for liquid-cooled servers based on the historical monitoring data set of liquid-cooled servers, the server temperature and temperature uniformity can be effectively predicted, and reasonable temperature adjustment and control measures can be made. This not only ensures the safety and stability of server operation, but also optimizes energy efficiency and reduces operating costs.

[0039] (2) Based on the predicted temperature and predicted temperature uniformity, a temperature control model for a liquid-cooled server is constructed and temperature control information is output, rather than considering temperature alone. The constructed model is more practical and subsequent regulation can be more accurate.

[0040] (3) By simplifying the constant terms, the model objective function is simplified and the temperature prediction model of the liquid-cooled server is obtained, which simplifies the calculation process and improves the system operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0042] Figure 1 The present invention is a flow chart of the steps of a temperature control prediction method for a liquid-cooled server. DETAILED DESCRIPTION

[0043] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0044] The working principle and use process of the present invention:

[0045] See also Figure 1 , a temperature control prediction method for a liquid cooling server, comprising:

[0046] S1: Obtain a historical monitoring data set of a liquid cooling server, wherein the historical monitoring data set of the liquid cooling server includes a historical temperature of the liquid cooling server, a historical operating computing power of the liquid cooling server, a historical operating power of the liquid cooling server, a historical operating primary oil pump speed of the liquid cooling server, a historical operating secondary water pump speed of the liquid cooling server, and a historical operating environment temperature of the liquid cooling server;

[0047] S2: constructing a temperature prediction model for a liquid cooling server based on the historical monitoring data set of the liquid cooling server by improving the XGBOOST algorithm;

[0048] S3: Obtaining a preset parameter set of a liquid cooling server, wherein the preset parameter set of the liquid cooling server includes the liquid cooling server operation computing power, the liquid cooling server operation power, the liquid cooling server operation primary oil pump speed, the liquid cooling server operation secondary water pump speed, and the liquid cooling server operation environment temperature;

[0049] S4: Inputting the preset parameter set of the liquid cooling server into the temperature prediction model of the liquid cooling server to obtain the predicted temperature and the predicted temperature uniformity, instead of considering the temperature alone, so that the constructed model is more practical and the subsequent regulation can be more accurate;

[0050] S5: Based on the predicted temperature and the predicted temperature uniformity, a liquid cooling server temperature control model is constructed and temperature control information is output, wherein the temperature control information includes temperature control abnormality information and temperature control normal information, and liquid cooling server parameter adjustment instructions are fed back based on the temperature control information, wherein the liquid cooling server parameter adjustment instructions include primary oil pump speed adjustment instructions, secondary water pump speed adjustment instructions, and operating computing power adjustment instructions.

[0051] In this embodiment, a liquid cooling server temperature prediction model is constructed based on the liquid cooling server historical monitoring data set by improving the XGBOOST algorithm, which can be implemented by the following steps:

[0052] S201: Acquire historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server, where the historical temperature of the liquid cooling server includes a CPU historical temperature, a GPU historical temperature, and a memory historical temperature;

[0053] S202: Given the historical monitoring data set of the liquid cooling server and the historical temperature uniformity of the liquid cooling server, the improved XGBOOST algorithm is used to perform additive learning of k decision trees to construct a temperature prediction model for the liquid cooling server, effectively predict the server temperature and temperature uniformity, and make reasonable temperature adjustment and control measures, which not only ensures the safety and stability of the server operation, but also optimizes energy utilization efficiency and reduces operating costs.

[0054] In this embodiment, obtaining the historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server can be specifically implemented by the following steps:

[0055] S201-1: Obtain the number of heat-generating components in the liquid cooling server, where the number of heat-generating components includes the number of CPUs, the number of GPUs, and the number of memories;

[0056] S201-2: Calculate the historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server and the number of the heat generating components. The calculation formula is: Among them, S is the historical temperature uniformity of the liquid cooling server, N1 is the number of CPUs, N2 is the number of GPUs, N3 is the number of memories, and T i1 is the CPU historical temperature of the i-th CPU, T ave1 is the historical average temperature of N1 CPUs, T i2 is the GPU historical temperature of the i-th GPU, T ave2 is the average historical temperature of N2 GPUs, T i3 is the memory history temperature of the i-th memory, T ave3 It is the historical average temperature of N3 memories.

[0057] In this embodiment, the improved XGBOOST algorithm is used to perform additive learning of k decision trees to construct the temperature prediction model of the liquid cooling server, which can be specifically implemented by the following steps:

[0058] S202-1: Obtain the loss function of the temperature prediction model of the liquid cooling server, wherein the loss function includes a temperature loss function and a temperature uniformity loss function. The expression of the temperature loss function is: Among them, l is the temperature loss function, y j is the historical temperature of the liquid-cooled server, is the predicted historical temperature output by the j-th decision tree of the model at the t-th iteration, is the predicted historical temperature output by the jth decision tree of the model at the t-1th iteration, is the predicted historical temperature of the jth decision tree in the model at the tth iteration, k is the number of decision trees in the model; the expression of the temperature uniformity loss function is Where w is the temperature uniformity loss function, r j For the historical temperature uniformity of liquid-cooled servers, is the predicted historical temperature uniformity output by the j-th decision tree of the model at the t-th iteration, is the predicted historical temperature uniformity output by the jth decision tree of the model at the t-1th iteration, is the predicted historical temperature uniformity of the jth decision tree in the model at the tth iteration, and k is the number of decision trees in the model;

[0059] S202-2: Obtain the model regularization term of the liquid cooling server temperature prediction model, the expression of the model regularization term is: Among them, θ is the model regularization term, α is the first quantization regularization term parameter, λ is the second quantization regularization term parameter, S is the number of model decision tree leaf nodes, T is the number of decision tree leaves, ω t Output the result for the tth leaf of the decision tree;

[0060] S202-3: Obtain a model objective function based on the loss function and the model regularization term. The expression of the model objective function is: Among them, R is the model objective function;

[0061] S202-4: By expanding the model objective function by a second-order Taylor expansion, we can obtain , where h i is the first-order derivative of the temperature loss function, p i is the second-order derivative of the temperature loss function, q i is the first-order derivative of the temperature uniformity loss function, z i is the second-order derivative of the temperature uniformity loss function;

[0062] S202-5: Simplify the model objective function by simplifying the constant term and obtain the liquid cooling server temperature prediction model, and obtain Simplify the calculation process and improve system operation efficiency.

[0063] In this embodiment, a liquid cooling server temperature control model is constructed based on the predicted temperature and the predicted temperature uniformity, and temperature control information is output, which can be specifically implemented by the following steps:

[0064] S501: Preset a predicted temperature threshold and a predicted temperature uniformity threshold. When the predicted temperature is greater than or equal to the predicted temperature threshold or the predicted temperature uniformity is less than or equal to the predicted temperature uniformity threshold, output the temperature control abnormality information. When the predicted temperature is less than the predicted temperature threshold and the predicted temperature uniformity is greater than the predicted temperature uniformity threshold, output the temperature control normal information.

[0065] In this embodiment, the liquid cooling server parameter adjustment instruction is fed back based on the temperature control information, which can be implemented by the following steps:

[0066] S502-1: when the temperature control abnormality information is output, determining the speed of the primary oil pump of the liquid cooling server, and when the speed of the primary oil pump of the liquid cooling server is less than the maximum speed of the primary oil pump, feeding back the primary oil pump speed adjustment instruction, and adjusting the speed of the primary oil pump of the liquid cooling server until the temperature control normal information is output;

[0067] S502-2: when the speed of the primary oil pump of the liquid-cooled server is equal to the maximum speed, determining the speed of the secondary water pump of the liquid-cooled server; when the speed of the secondary water pump of the liquid-cooled server is less than the maximum speed of the secondary water pump, feeding back the secondary water pump speed adjustment instruction, and adjusting the speed of the secondary water pump of the liquid-cooled server until the temperature control normal information is output;

[0068] S502-3: When the speed of the secondary water pump of the liquid cooling server is equal to the maximum speed, the operation computing power adjustment instruction is fed back to reduce the operation computing power of the liquid cooling server until the temperature control is normal information is output.

[0069] A temperature control prediction device for a liquid cooling server, comprising a data acquisition module, a model building module, and a temperature control module;

[0070] The data acquisition module is used to acquire a historical monitoring data set of a liquid cooling server, and the historical monitoring data set of the liquid cooling server includes a historical temperature of the liquid cooling server, a historical operating computing power of the liquid cooling server, a historical operating power of the liquid cooling server, a historical operating primary oil pump speed of the liquid cooling server, a historical operating secondary water pump speed of the liquid cooling server, and a historical operating environment temperature of the liquid cooling server;

[0071] The model building module is used to build a liquid cooling server temperature prediction model based on the liquid cooling server historical monitoring data set by improving the XGBOOST algorithm; obtain a liquid cooling server preset parameter set, the liquid cooling server preset parameter set includes the liquid cooling server operation computing power, the liquid cooling server operation power, the liquid cooling server operation primary oil pump speed, the liquid cooling server operation secondary water pump speed, and the liquid cooling server operation environment temperature; the liquid cooling server preset parameter set is input into the liquid cooling server temperature prediction model to obtain the predicted temperature and the predicted temperature uniformity;

[0072] The temperature control module is used to build a liquid cooling server temperature control model based on the predicted temperature and the predicted temperature uniformity and output temperature control information, wherein the temperature control information includes temperature control abnormality information and temperature control normal information, and to feedback liquid cooling server parameter adjustment instructions based on the temperature control information, wherein the liquid cooling server parameter adjustment instructions include primary oil pump speed adjustment instructions, secondary water pump speed adjustment instructions, and operating computing power adjustment instructions.

[0073] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0074] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0075] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0076] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A temperature control prediction method for a liquid cooling server, characterized in that: The implementation of the temperature control prediction method comprises the following steps: Acquire a historical monitoring data set of a liquid cooling server, wherein the historical monitoring data set of the liquid cooling server includes a historical temperature of the liquid cooling server, a historical operating computing power of the liquid cooling server, a historical operating power of the liquid cooling server, a historical operating primary oil pump speed of the liquid cooling server, a historical operating secondary water pump speed of the liquid cooling server, and a historical operating environment temperature of the liquid cooling server; Building a temperature prediction model for a liquid cooling server based on the historical monitoring data set of the liquid cooling server by improving the XGBOOST algorithm; Obtaining a preset parameter set of a liquid cooling server, wherein the preset parameter set of the liquid cooling server includes a liquid cooling server operating computing power, a liquid cooling server operating power, a liquid cooling server operating primary oil pump speed, a liquid cooling server operating secondary water pump speed, and a liquid cooling server operating environment temperature; Inputting the preset parameter set of the liquid cooling server into the temperature prediction model of the liquid cooling server to obtain the predicted temperature and the predicted temperature uniformity; Based on the predicted temperature and the predicted temperature uniformity, a liquid cooling server temperature control model is constructed and temperature control information is output, wherein the temperature control information includes temperature control abnormality information and temperature control normal information. Based on the temperature control information, liquid cooling server parameter adjustment instructions are fed back, wherein the liquid cooling server parameter adjustment instructions include primary oil pump speed adjustment instructions, secondary water pump speed adjustment instructions, and operating computing power adjustment instructions.

2. The temperature control prediction method according to claim 1, characterized in that: The method of constructing a temperature prediction model for a liquid cooling server based on the historical monitoring data set of the liquid cooling server by improving the XGBOOST algorithm includes: Acquire the historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server, wherein the historical temperature of the liquid cooling server includes the historical temperature of the CPU, the historical temperature of the GPU, and the historical temperature of the memory; Given the historical monitoring data set of the liquid cooling server and the historical temperature uniformity of the liquid cooling server, the improved XGBOOST algorithm is used to perform additive learning of k decision trees to construct the temperature prediction model of the liquid cooling server.

3. The temperature control prediction method according to claim 2, characterized in that: The obtaining of the historical temperature uniformity of the liquid cooling server based on the historical temperature of the liquid cooling server includes: Obtain the number of heat-generating components in the liquid-cooled server, where the number of heat-generating components includes the number of CPUs, the number of GPUs, and the number of memories; The historical temperature uniformity of the liquid cooling server is calculated based on the historical temperature of the liquid cooling server and the number of the heat generating components.

4. The temperature control prediction method according to claim 2, characterized in that: The method of performing additive learning of k decision trees by the improved XGBOOST algorithm to construct the temperature prediction model of the liquid cooling server includes: Obtaining a loss function of the temperature prediction model of the liquid cooling server, wherein the loss function includes a temperature loss function and a temperature uniformity loss function; Obtaining a model regularization term of the liquid cooling server temperature prediction model; Obtaining a model objective function based on the loss function and the model regularization term; Expand the model objective function by a second-order Taylor expansion; The model objective function is simplified by simplifying constant terms and the liquid cooling server temperature prediction model is obtained.

5. The temperature control prediction method according to claim 4, characterized in that: The expression of the temperature loss function is: Among them, l is the temperature loss function, y j is the historical temperature of the liquid-cooled server, is the predicted historical temperature output by the j-th decision tree of the model at the t-th iteration, is the predicted historical temperature output by the jth decision tree of the model at the t-1th iteration, is the predicted historical temperature of the jth decision tree in the model at the tth iteration, k is the number of decision trees in the model; the expression of the temperature uniformity loss function is Where w is the temperature uniformity loss function, r j For the historical temperature uniformity of liquid-cooled servers, is the predicted historical temperature uniformity output by the j-th decision tree of the model at the t-th iteration, is the predicted historical temperature uniformity output by the jth decision tree of the model at the t-1th iteration, is the predicted historical temperature uniformity of the jth decision tree in the model at the tth iteration.

6. The temperature control prediction method according to claim 1, characterized in that: The step of constructing a temperature control model for a liquid cooling server based on the predicted temperature and the predicted temperature uniformity and outputting temperature control information comprises: A predicted temperature threshold and a predicted temperature uniformity threshold are preset. When the predicted temperature is greater than or equal to the predicted temperature threshold or the predicted temperature uniformity is less than or equal to the predicted temperature uniformity threshold, the temperature control abnormality information is output; when the predicted temperature is less than the predicted temperature threshold and the predicted temperature uniformity is greater than the predicted temperature uniformity threshold, the temperature control normal information is output.

7. The temperature control prediction method according to claim 1, characterized in that: The liquid cooling server parameter adjustment instruction based on the temperature control information feedback includes: When the temperature control abnormality information is output, the speed of the oil pump of the liquid cooling server is judged, and when the speed of the oil pump of the liquid cooling server is less than the maximum speed of the oil pump, the speed adjustment instruction of the oil pump is fed back to adjust the speed of the oil pump of the liquid cooling server until the normal temperature control information is output; When the speed of the primary oil pump of the liquid-cooled server is equal to the maximum speed, the speed of the secondary water pump of the liquid-cooled server is determined; when the speed of the secondary water pump of the liquid-cooled server is less than the maximum speed of the secondary water pump, the speed adjustment instruction of the secondary water pump is fed back to adjust the speed of the secondary water pump of the liquid-cooled server until the normal temperature control information is output; When the speed of the secondary water pump of the liquid cooling server is equal to the maximum speed, the operation computing power adjustment instruction is fed back to reduce the operation computing power of the liquid cooling server until the temperature control normal information is output.

8. A temperature control prediction device for a liquid cooling server, characterized in that: The device is applied to the temperature control prediction method as described in any one of claims 1 to 7, comprising a data acquisition module, a model building module, and a temperature control module; The data acquisition module is used to acquire a historical monitoring data set of a liquid cooling server, and the historical monitoring data set of the liquid cooling server includes a historical temperature of the liquid cooling server, a historical operating computing power of the liquid cooling server, a historical operating power of the liquid cooling server, a historical operating primary oil pump speed of the liquid cooling server, a historical operating secondary water pump speed of the liquid cooling server, and a historical operating environment temperature of the liquid cooling server; The model building module is used to build a liquid cooling server temperature prediction model based on the liquid cooling server historical monitoring data set by improving the XGBOOST algorithm; obtain a liquid cooling server preset parameter set, the liquid cooling server preset parameter set includes the liquid cooling server operation computing power, the liquid cooling server operation power, the liquid cooling server operation primary oil pump speed, the liquid cooling server operation secondary water pump speed, and the liquid cooling server operation environment temperature; the liquid cooling server preset parameter set is input into the liquid cooling server temperature prediction model to obtain the predicted temperature and the predicted temperature uniformity; The temperature control module is used to build a liquid cooling server temperature control model based on the predicted temperature and the predicted temperature uniformity and output temperature control information, wherein the temperature control information includes temperature control abnormality information and temperature control normal information, and to feedback liquid cooling server parameter adjustment instructions based on the temperature control information, wherein the liquid cooling server parameter adjustment instructions include primary oil pump speed adjustment instructions, secondary water pump speed adjustment instructions, and operating computing power adjustment instructions.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the temperature control prediction method as described in any one of claims 1-7 is implemented.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to perform the temperature control prediction method according to any one of claims 1 to 7 when executed by a computer processor.

Images