Method, device, computer device and storage medium for performing heat dissipation operation

Abstract

The application relates to the technical field of servers and discloses a method and device for performing heat dissipation operation, computer equipment and a storage medium, which comprise the following steps: in a current period, the actual temperature values of each heat generating component in multiple periods are acquired; the actual temperature values of each heat generating component in multiple periods are input into a pre-constructed temperature prediction model to obtain the predicted temperature values of each heat generating component in a next period; whether heat dissipation operation is performed on a target heat generating component is determined according to the actual temperature value of the target heat generating component in the current period and the predicted temperature value of the target heat generating component in the next period; when it is determined that heat dissipation operation is performed on the target heat generating component, the actual temperature value of the target heat generating component in the current period is input into a pre-constructed heat dissipation control model to obtain a target control amount; and the heat dissipation operation amount of a target heat dissipating component within a preset distance range of the target heat generating component is improved according to the target control amount. The application can reduce resource waste.

Description

Technical Field

[0001] This invention relates to the field of server technology, and more specifically to methods, apparatus, computer equipment, and storage media for performing heat dissipation operations. Background Technology

[0002] In the field of server technology, servers are a crucial component of modern information technology infrastructure. Because servers typically handle heavy computing tasks, they generate a significant amount of heat. If this heat cannot be dissipated effectively and promptly, it can lead to overheating, affecting performance and even causing hardware damage. Therefore, servers usually require various heat dissipation components to effectively cool them. For example, a fan can be a heat dissipation component.

[0003] In related technologies, servers typically collect their own temperature values ​​periodically, compare the temperature value collected in the current period with a preset temperature value, and if the temperature value collected in the current period is greater than the preset temperature value, then a control quantity is calculated based on the temperature value collected in the current period and the control model, and then the fan speed is increased according to the control quantity to reduce the temperature.

[0004] However, the related technologies can lead to problems of excessive or insufficient heat dissipation, resulting in a waste of resources. Summary of the Invention

[0005] In view of this, the present invention provides a method, apparatus, computer device, storage medium and program product for performing heat dissipation operations to solve the problem that related technologies may lead to excessive heat dissipation or insufficient heat dissipation.

[0006] In a first aspect, the present invention provides a method for performing a heat dissipation operation, the method being applied to a server, the server including at least one heat-generating component, the method comprising:

[0007] In the current cycle, the actual temperature value of each of the heating components is obtained in multiple cycles, wherein the multiple cycles include the current cycle and at least one cycle before the current cycle;

[0008] The actual temperature values ​​of each of the heating components in multiple cycles are input into a pre-built temperature prediction model to obtain the predicted temperature value of each of the heating components in the next cycle corresponding to the current cycle.

[0009] Based on the actual temperature value of the target heating component in the current cycle and the predicted temperature value in the next cycle, it is determined whether to perform a heat dissipation operation on the target heating component, wherein the target heating component is any one of at least one of the heating components;

[0010] When it is determined that a heat dissipation operation will be performed on the target heat-generating component, the actual temperature value of the target heat-generating component in the current cycle is input into the pre-built heat dissipation control model to obtain the target control quantity;

[0011] Based on the target control quantity, the heat dissipation operation quantity of the target heat dissipation component within the preset distance range of the target heat-generating component is increased.

[0012] The present invention provides a method for performing heat dissipation operations, which has the following advantages:

[0013] This solution periodically collects the actual temperature value of each heat-generating component. In the current cycle, the server can predict the temperature value of each heat-generating component in the next cycle based on its actual temperature values ​​in the current cycle and a series of previous cycles. Furthermore, for the target heat-generating component, the server can determine its future trend—whether the temperature will rise or fall—by comparing its actual and predicted temperature values ​​in the current cycle, and then execute heat dissipation operations based on the actual temperature value of the current cycle. This way, if the future temperature trend is upward, the server can preemptively dissipate heat, avoiding the need for higher power cooling when more heat accumulates later, thus saving resources. If the future temperature trend is downward, performing heat dissipation operations in the current cycle would lead to excessive heat dissipation and resource waste. Additionally, this solution uses the actual temperature value of the target heat-generating component in the current cycle to calculate the target control variable, rather than the predicted temperature value. Therefore, when heat dissipation operations are performed, using the actual temperature value of the current cycle saves resources.

[0014] In one optional implementation, the temperature prediction model includes multiple weight matrices, multiple activation functions, and multiple bias vectors. The weight matrices are used to perform linear transformations on the input vectors, the activation functions are used to perform nonlinear transformations on the input vectors, and the bias vectors are used to adjust the activation level of their corresponding activation functions.

[0015] The step of inputting the actual temperature values ​​of each of the heating components in multiple cycles into a pre-built temperature prediction model to obtain the predicted temperature value of each of the heating components in the next cycle corresponding to the current cycle, as output by the temperature prediction model, includes:

[0016] Based on the actual temperature value of each of the heating components in the first cycle, a first data input vector corresponding to the first cycle is generated, wherein the first cycle is any cycle other than the first cycle among the plurality of cycles;

[0017] The second data output vector generated in the second cycle is concatenated with the first data input vector to obtain the total data input vector corresponding to the first cycle, wherein the second cycle is the previous cycle of the first cycle;

[0018] Each activation function is used as a target activation function. The total data input vector corresponding to the first period, the target weight matrix corresponding to the target activation function, and the target bias vector are input into the target activation function to obtain the transformation vector output by the target activation function.

[0019] Obtain the first state vector corresponding to the second cycle;

[0020] Based on the transformation vector output by each activation function and the first state vector, a first data output vector for the first period is generated;

[0021] After generating a data output vector corresponding to the last cycle among the plurality of cycles, the data output vector corresponding to the last cycle is transformed in dimension according to the first weight matrix in at least one of the weight matrices to obtain a predicted temperature vector, wherein the predicted temperature vector includes the predicted temperature value of each of the heating components in the next cycle.

[0022] Specifically, during the prediction process, the temperature prediction model can combine previous temperatures with the current cycle's temperature to predict the temperature of the next cycle, enabling it to predict more accurate temperature values ​​by analyzing temperature evolution trends. Furthermore, by processing the total data input vector multiple times using multiple weight matrices, activation functions, and bias vectors, the temperature prediction model can learn the characteristics of temperature changes and predict more accurate temperature values. In summary, this approach provides a more accurate basis for subsequent heat dissipation operations, avoiding resource waste caused by erroneous heat dissipation actions.

[0023] In one optional implementation, generating the first data output vector for the first period based on the transformation vector output by each of the activation functions and the first state vector includes:

[0024] Based on the transformation vectors output by the activation functions other than the preset activation function and the first state vector, determine the second state vector corresponding to the first period;

[0025] The first data output vector is determined based on the transformation vector output by the preset activation function and the second state vector.

[0026] Specifically, by combining the transformation vector and state vector output by the activation function, the temperature prediction model can learn the characteristics of temperature changes and predict more accurate temperature values.

[0027] In one optional implementation, each activation function is used as the target activation function. The total data input vector corresponding to the first cycle, the target weight matrix corresponding to the target activation function, and the target bias vector are input into the target activation function to obtain the transformation vector output by the target activation function, using the following expression:

[0028]

[0029] in, Let J be the transformation vector output by the target activation function, which can be either the sigmoid function or the hyperbolic tangent function. This is the total data input vector for the first period. This is the first data input vector for the first cycle. This is the second data output vector for the second cycle. For the target bias vector, This is the target weight matrix.

[0030] In one optional implementation, the step of determining the second state vector corresponding to the first period based on the transformation vectors output by the activation functions other than the preset activation function and the first state vector is expressed by the following expression:

[0031]

[0032] in, This is the second state vector. The transformation vector output by the first activation function among the plurality of activation functions. Let this be the first state vector. The transformation vector is the output of the second activation function among the plurality of activation functions. The transformation vector is the output of the third activation function among the multiple activation functions.

[0033] In one optional implementation, the first data output vector is determined based on the transformation vector output by the preset activation function and the second state vector, using the following expression:

[0034]

[0035] in, Output vector for the first data. The transformation vector output by the preset activation function. This is the second state vector.

[0036] In one optional implementation, determining whether to perform a heat dissipation operation on the target heat-generating component based on the actual temperature value of the target heat-generating component in the current cycle and the predicted temperature value in the next cycle includes:

[0037] The temperature change value of the target heating element is obtained by subtracting the predicted temperature value of the target heating element in the next cycle from the actual temperature value of the target heating element in the current cycle.

[0038] Determine whether the temperature change value of the target heating element is greater than or equal to a preset temperature change threshold;

[0039] When it is determined that the temperature change value of the target heating component is greater than or equal to the temperature change threshold, it is determined that a heat dissipation operation will be performed on the target heating component.

[0040] or,

[0041] When it is determined that the temperature change value of the target heat-generating component is less than the temperature change threshold, it is determined that no heat dissipation operation will be performed on the target heat-generating component.

[0042] Specifically, by comparing the temperature change value with the temperature change threshold, the server can start taking heat dissipation measures before it reaches a dangerous level, which can effectively prevent heat-generating components from overheating and protect the server's safety.

[0043] In a second aspect, the present invention provides an apparatus for performing a heat dissipation operation, the apparatus being applied to a server, the server including at least one heat-generating component, the apparatus comprising:

[0044] In the current cycle, the actual temperature value of each of the heating components is obtained in multiple cycles, wherein the multiple cycles include the current cycle and at least one cycle before the current cycle;

[0045] The actual temperature values ​​of each of the heating components in multiple cycles are input into a pre-built temperature prediction model to obtain the predicted temperature value of each of the heating components in the next cycle corresponding to the current cycle.

[0046] Based on the actual temperature value of the target heating component in the current cycle and the predicted temperature value in the next cycle, it is determined whether to perform a heat dissipation operation on the target heating component, wherein the target heating component is any one of at least one of the heating components;

[0047] When it is determined that a heat dissipation operation will be performed on the target heat-generating component, the actual temperature value of the target heat-generating component in the current cycle is input into the pre-built heat dissipation control model to obtain the target control quantity;

[0048] Based on the target control quantity, the heat dissipation operation quantity of the target heat dissipation component within the preset distance range of the target heat-generating component is increased.

[0049] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method for performing heat dissipation operation as described in the first aspect or any corresponding embodiment.

[0050] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the method for performing heat dissipation operations as described in the first aspect or any corresponding embodiment thereof.

[0051] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to perform a method for performing heat dissipation operations as described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the specific embodiments or related technologies of the present invention, the drawings used in the description of the specific embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0053] Figure 1 This is a schematic diagram of the server architecture according to an embodiment of the present invention;

[0054] Figure 2 This is a flowchart illustrating a method for performing heat dissipation operations according to an embodiment of the present invention;

[0055] Figure 3 This is a structural block diagram of a device for performing heat dissipation operation according to an embodiment of the present invention;

[0056] Figure 4 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] This invention provides a server, such as Figure 1As shown, the server can include various heat-generating components that produce heat during operation, such as a Central Processing Unit (CPU), Graphics Processing Unit (GPU), memory modules, and hard drives. The surfaces of these heat-generating components can be coated with graphene to improve their heat dissipation efficiency. To collect the temperature of these components, each can integrate a temperature sensor, such as a Negative Temperature Coefficient (NTC) resistor or a Resistance Temperature Detector (RTD). To dissipate heat from these components, the server can also include heat dissipation components, such as fans, pumps, solenoid valves, and coolant pipes. Additionally, the server can include at least one thermoelectric converter, which absorbs the heat generated inside the server and converts it into electrical energy. The thermoelectric converter operates on the Seebeck effect, i.e., V = SΔT, where V is the voltage, S is the Seebeck coefficient, and ΔT is the temperature difference. In some alternative implementations, the server structure can be obtained using Finite Element Analysis (FEA) software. Specifically, since the CPU performs a significant number of computational tasks and is one of the most important components of the server, a pre-set CPU temperature reduction threshold can be established, for example, 10%. Technicians can pre-configure a 3D geometric model of the server's interior, use FEA software to mesh the model, and define the material properties of heat-generating components, such as thermal conductivity and specific heat capacity, as well as define boundary conditions, the CPU's maximum power consumption, ambient temperature range, and the operating status of the cooling system. The operating status of the cooling system can include the fan speed range under different loads, the flow rate, pressure, and initial temperature of the liquid in the liquid cooling system, etc. Furthermore, by changing the shape, number, and distribution of the heat dissipation components inside the server, the FEA software can simulate the CPU's temperature change rate under different server structures. Based on the CPU's heat dissipation under different server structures, the server structure with the largest temperature change rate is selected. During the simulation, the FEA software can use the Fourier heat conduction formula: Where T represents the temperature value, t represents the preset thermal diffusivity and time.

[0059] This invention provides a method for performing heat dissipation operations. By using the actual temperature values ​​detected in the current cycle and those detected before the current cycle, the temperature value of the next cycle is predicted. This allows for the determination of subsequent temperature trends and the early execution of heat dissipation operations, thereby saving resources.

[0060] According to an embodiment of the present invention, a method embodiment for performing heat dissipation operation is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0061] This embodiment provides a method for performing heat dissipation operations, which can be used in the aforementioned server. Figure 2 This is a flowchart of a method for performing heat dissipation operation according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0062] Step S201: In the current cycle, obtain the actual temperature value of each heating component in multiple cycles.

[0063] Among these multiple cycles, the current cycle and at least one cycle preceding the current cycle may be included.

[0064] Specifically, the server can periodically collect and store the actual temperature values ​​of the corresponding heat-generating components from each temperature sensor according to a pre-configured cycle duration. After completing the collection operation for the current cycle, the server can predict the temperature value of each heat-generating component in the next cycle based on the actual temperature values ​​of each heat-generating component collected in the current cycle and at least one cycle prior to the current cycle.

[0065] Step S202: Input the actual temperature values ​​of each heating component in multiple cycles into the pre-built temperature prediction model to obtain the predicted temperature value of each heating component in the next cycle corresponding to the current cycle.

[0066] Specifically, the temperature prediction model can include multiple weight matrices, multiple activation functions, and multiple bias vectors. The temperature prediction model can be a Long Short-Term Memory (LSTM) network. The weight matrices can be used to perform linear transformations on the input vectors, the activation functions can be used to perform nonlinear transformations on the input vectors, and the bias vectors can be used to adjust the activation level of their corresponding activation functions. Accordingly, the process of the temperature prediction model calculating the predicted temperature value can be as follows:

[0067] Step 1: Generate the first data input vector corresponding to the first cycle based on the actual temperature value of each heating component in the first cycle.

[0068] The first cycle refers to any cycle other than the first cycle among multiple cycles.

[0069] For example, there are 3 heating elements. In the first cycle, the actual temperature values ​​of the three heating elements are 25, 30 and 35 respectively, and the first data input vector is [25, 20, 35].

[0070] Step two: Concatenate the second data output vector generated in the second cycle with the first data input vector to obtain the total data input vector corresponding to the first cycle.

[0071] The second cycle is the cycle preceding the first cycle.

[0072] Step 3: Using each activation function as the target activation function, input the total data input vector corresponding to the first cycle, the target weight matrix corresponding to the target activation function, and the target bias vector into the target activation function to obtain the transformation vector output by the target activation function.

[0073]

[0074] in, Let J be the transformation vector output by the target activation function, which can be either the sigmoid function or the hyperbolic tangent function. This is the total data input vector for the first period. This is the first data input vector for the first cycle. This is the second data output vector for the second cycle. For the target bias vector, This is the target weight matrix.

[0075] In some optional implementations, the multiple activation functions may include a first activation function, a second activation function, a third activation function, and a fourth activation function, wherein the first activation function, the second activation function, and the fourth activation function may be a sigmoid function, which can be expressed as: The third activation function can be the hyperbolic tangent function. For example, the transformation vector output by the first activation function can be as follows:

[0076]

[0077] in, It is the transformation vector output by the first activation function. (x) is the first activation function, and x is , It is the second weight matrix. It is the first bias vector.

[0078] The transformation vector output by the second activation function can be as follows:

[0079]

[0080] in, It is the transformation vector output by the second activation function. It is the third weight matrix. It is the second bias vector.

[0081] The transformation vector output by the third activation function can be as follows:

[0082]

[0083] in, It is the transformation vector output by the third activation function. It is the fourth weight matrix. It is the third bias vector.

[0084] The transformation vector output by the preset activation function can be as follows:

[0085]

[0086] in, It is the transformation vector output by the preset activation function. It is the fifth weight matrix. It is the fourth bias vector.

[0087] Step 4: Obtain the first state vector corresponding to the second cycle.

[0088] Step 5: Generate the first data output vector for the first period based on the transformation vector and the first state vector output by each activation function.

[0089] Step 1: Determine the second state vector corresponding to the first period based on the transformation vector and the first state vector output by the activation functions other than the preset activation function.

[0090] Specifically, step 1 can be expressed as follows:

[0091]

[0092] The first state vector corresponding to the second cycle can be calculated using this formula during the processing of the second cycle.

[0093] Step 2: Determine the first data output vector based on the transformation vector and the second state vector output by the preset activation function.

[0094] Specifically, step 2 can be expressed as follows:

[0095]

[0096] in, It is the first data output vector.

[0097] Step 6: After generating the data output vector corresponding to the last period among multiple periods, perform a dimension transformation on the data output vector corresponding to the last period according to the first weight matrix in at least one weight matrix to obtain the predicted temperature vector.

[0098] The predicted temperature vector includes the predicted temperature value of each heating component in the next cycle.

[0099] The calculation of the data output vector for the first cycle is similar to the above calculation process, except that in step two, the server can concatenate the zero vector of the first preset dimension with the data input vector corresponding to the first cycle to obtain the total data input vector corresponding to the first cycle, as described above. It can be the zero vector of the first preset dimension. When calculating the state vector of the first cycle, the zero vector of the second preset dimension can be obtained in step four, i.e., the one described above. It can be the zero vector of the second preset dimension. In step five, the second state vector of the first cycle can be determined based on the transformation vectors and the zero vector of the second preset dimension output by the activation functions other than the preset activation function.

[0100] Step S203: Based on the actual temperature value of the target heat-generating component in the current cycle and the predicted temperature value in the next cycle, determine whether to perform a heat dissipation operation on the target heat-generating component.

[0101] The target heating element is any one of the at least one heating elements.

[0102] Specifically, step S203 may include the following steps:

[0103] Step 1: Subtract the actual temperature of the target heating element in the current cycle from the predicted temperature value of the target heating element in the next cycle to obtain the temperature change value of the target heating element.

[0104] Step 2: Determine whether the temperature change value of the target heating component is greater than or equal to the preset temperature change threshold.

[0105] Step 3: When the temperature change value of the target heat-generating component is determined to be greater than or equal to the temperature change threshold, a heat dissipation operation is performed on the target heat-generating component.

[0106] Step 4: When it is determined that the temperature change value of the target heat-generating component is less than the temperature change threshold, it is determined that no heat dissipation operation will be performed on the target heat-generating component.

[0107] In some optional implementations, if the server determines that the temperature change value of the target heat-generating component is less than a temperature change threshold and the temperature change threshold is a value less than 0, the server may determine not to perform a heat dissipation operation on the target heat-generating component. Alternatively, if the server determines that the temperature change value of the target heat-generating component is less than a temperature change threshold and the temperature change threshold is a value greater than or equal to 0, the server may also determine whether the predicted temperature value of the target heat-generating component is greater than or equal to a preset temperature threshold. If the server determines that the predicted temperature value of the target heat-generating component is greater than the preset temperature threshold, the server may also perform a heat dissipation operation on the target heat-generating component. Alternatively, if the server determines that the predicted temperature value of the target heat-generating component is less than or equal to the preset temperature threshold, the server may then determine not to perform a heat dissipation operation on the target heat-generating component.

[0108] In this way, by further considering the relationship between the predicted temperature value and the preset temperature threshold, it is possible to more accurately determine whether the predicted temperature value is safe and avoid safety problems caused by a large predicted temperature value.

[0109] Step S204: When it is determined that a heat dissipation operation will be performed on the target heat-generating component, the actual temperature value of the target heat-generating component in the current cycle is input into the pre-built heat dissipation control model to obtain the target control quantity.

[0110] Specifically, step S204 can be expressed as follows:

[0111]

[0112] in, For the target control quantity, The preset proportional gain coefficient, The preset integral gain coefficient, To preset the differential gain coefficient, The temperature error value of the target heating component in the current cycle. Any period between the initial period and the current period The temperature error value is the difference between the actual temperature value and the preset safe temperature value.

[0113] Step S205: Based on the target control quantity, increase the heat dissipation operation quantity of the target heat dissipation component within the preset distance range of the target heat-generating component.

[0114] Specifically, when the server adopts a cooling method including fan cooling, and the target cooling component is the fan, the cooling operation quantity can be the rotation speed. When the server adopts a cooling method including liquid cooling, the target cooling components include pumps and solenoid valves. The cooling operation quantity corresponding to the solenoid valve can be the valve opening degree, and the cooling operation quantity corresponding to the pump can be the rotation speed.

[0115] At least one heat dissipation component can be installed within a preset distance range of each heat-generating component. For each heat-generating component, the server can obtain the identification information of the heat-generating component, and based on the identification information of the heat-generating component, obtain the identification information of at least one heat dissipation component within the preset distance range of the heat-generating component. Based on the identification information of the heat dissipation component and the target control quantity, the server can increase the heat dissipation operation of the corresponding heat dissipation component, for example, by increasing the fan speed, or by increasing the pump speed and the opening degree of the solenoid valve.

[0116] For example, if the actual temperature of the target heat-generating component in the current cycle is 70°C and the predicted temperature in the next cycle is 75°C, the target control quantity calculated by the heat dissipation control model is 3000 RPM (speed unit). The server can increase the speed of the fan corresponding to the target heat-generating component based on the target control quantity.

[0117] In some optional implementations, when it is determined that no heat dissipation operation will be performed on the target heat-generating component, the actual temperature value of the target heat-generating component in the current cycle is input into a pre-built heat dissipation control model to obtain a target control quantity. Based on the target control quantity, the heat dissipation operation quantity of the target heat-generating component within a preset distance range of the target heat-generating component is reduced. This avoids the problem of resource waste.

[0118] The method for performing heat dissipation operations provided in this embodiment can periodically collect the actual temperature value of each heat-generating component. Taking the current cycle as an example, the server can predict the temperature value of each heat-generating component in the next cycle based on its actual temperature values ​​in the current cycle and a series of previous cycles. Furthermore, for the target heat-generating component, the server can determine its future trend—whether the temperature will rise or fall—by using its actual temperature value in the current cycle and the predicted temperature value. Then, heat dissipation operations are performed based on the actual temperature value of the current cycle. In this way, if the future temperature trend is upward, the server can preemptively dissipate heat, avoiding the need for higher power cooling when more heat accumulates later, thus saving resources. If the future temperature trend is downward, performing heat dissipation operations in the current cycle would lead to excessive heat dissipation and resource waste. Additionally, this solution uses the actual temperature value of the target heat-generating component in the current cycle to calculate the target control quantity, rather than the predicted temperature value. Therefore, when performing heat dissipation operations, using the actual temperature value of the current cycle saves resources.

[0119] In summary, using thermally conductive materials and a server structure with good thermal conductivity for heat-generating components, and simultaneously employing both fans and liquid cooling systems to dissipate heat from the server, can improve the server's heat dissipation efficiency.

[0120] This embodiment also provides a device for performing heat dissipation operations, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0121] This embodiment provides a device for performing heat dissipation operations, such as... Figure 3 As shown, it includes:

[0122] The acquisition module 301 is used to acquire the actual temperature value of each heating component in multiple cycles in the current cycle, wherein the multiple cycles include the current cycle and at least one cycle before the current cycle;

[0123] The determination module 302 is used to input the actual temperature values ​​of each heating component in multiple cycles into a pre-built temperature prediction model to obtain the predicted temperature value of each heating component in the next cycle corresponding to the current cycle. Based on the actual temperature value of the target heating component in the current cycle and the predicted temperature value in the next cycle, it determines whether to perform a heat dissipation operation on the target heating component, wherein the target heating component is any one of at least one heating component.

[0124] The enhancement module 303 is used to input the actual temperature value of the target heat-generating component in the current cycle into the pre-built heat dissipation control model when it is determined that a heat dissipation operation will be performed on the target heat-generating component, so as to obtain the target control quantity; and to enhance the heat dissipation operation quantity of the target heat-generating component within a preset distance range of the target heat-generating component based on the target control quantity.

[0125] In some optional implementations, the temperature prediction model includes multiple weight matrices, multiple activation functions, and multiple bias vectors. The weight matrices are used to perform linear transformations on the input vectors, the activation functions are used to perform nonlinear transformations on the input vectors, and the bias vectors are used to adjust the activation level of their corresponding activation functions.

[0126] Module 302 is specifically used for:

[0127] Based on the actual temperature value of each heating component in the first cycle, a first data input vector corresponding to the first cycle is generated, wherein the first cycle is any cycle other than the first cycle among multiple cycles;

[0128] The second data output vector generated in the second cycle is concatenated with the first data input vector to obtain the total data input vector corresponding to the first cycle, where the second cycle is the cycle preceding the first cycle;

[0129] Each activation function is used as the target activation function. The total data input vector corresponding to the first cycle, the target weight matrix corresponding to the target activation function, and the target bias vector are input into the target activation function to obtain the transformation vector output by the target activation function.

[0130] Obtain the first state vector corresponding to the second cycle;

[0131] Based on the transformation vector and the first state vector output by each activation function, the first data output vector for the first period is generated;

[0132] After generating the data output vector corresponding to the last cycle among multiple cycles, the data output vector corresponding to the last cycle is transformed in dimension according to the first weight matrix in at least one weight matrix to obtain the predicted temperature vector, wherein the predicted temperature vector includes the predicted temperature value of each heating component in the next cycle.

[0133] In some alternative implementations, the determining module 302 is specifically used for:

[0134] Based on the transformation vector and the first state vector output by the activation functions other than the preset activation function, determine the second state vector corresponding to the first period.

[0135] The first data output vector is determined based on the transformation vector and the second state vector output by the preset activation function.

[0136] In some optional implementations, each activation function is used as the target activation function. The total data input vector corresponding to the first period, the target weight matrix corresponding to the target activation function, and the target bias vector are input into the target activation function to obtain the transformation vector output by the target activation function, using the following expression:

[0137]

[0138] in, Let J be the transformation vector output by the target activation function, which can be either the sigmoid function or the hyperbolic tangent function. This is the total data input vector for the first period. This is the first data input vector for the first cycle. This is the second data output vector for the second cycle. For the target bias vector, This is the target weight matrix.

[0139] In some optional implementations, the second state vector corresponding to the first period is determined based on the transformation vector and the first state vector output by the activation functions other than the preset activation function, using the following expression:

[0140]

[0141] in, This is the second state vector. This is the transformation vector output by the first activation function among multiple activation functions. Let be the first state vector. The transformation vector is the output of the second activation function out of a set of activation functions. This is the transformation vector output by the third activation function out of a set of activation functions.

[0142] In some optional implementations, the first data output vector is determined based on the transformation vector and the second state vector output by the preset activation function, using the following expression:

[0143]

[0144] in, The first data output vector, The transformation vector output by the preset activation function. This is the second state vector.

[0145] In some alternative implementations, the determining module 302 is specifically used for:

[0146] The temperature change of the target heating element is obtained by subtracting the predicted temperature value of the target heating element in the next cycle from the actual temperature value of the target heating element in the current cycle.

[0147] Determine whether the temperature change value of the target heating element is greater than or equal to the preset temperature change threshold;

[0148] When the temperature change value of the target heat-generating component is determined to be greater than or equal to the temperature change threshold, a heat dissipation operation is performed on the target heat-generating component.

[0149] or,

[0150] When the temperature change of the target heat-generating component is determined to be less than the temperature change threshold, it is determined that no heat dissipation operation will be performed on the target heat-generating component. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0151] In this embodiment, the device for performing heat dissipation is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0152] This invention also provides a computer device having the above-described features. Figure 3 The device shown performs the heat dissipation operation.

[0153] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 4 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 10 as an example.

[0154] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware integrated circuit. The aforementioned hardware integrated circuit may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The aforementioned programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0155] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0156] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0157] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0158] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0159] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0160] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0161] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for performing a heat dissipation operation, characterized in that, The method is applied to a server, the server including at least one heat-generating component, the method comprising: In the current cycle, the actual temperature value of each of the heating components is obtained in multiple cycles, wherein the multiple cycles include the current cycle and at least one cycle before the current cycle; Based on the actual temperature value of each heating component in the first cycle, a first data input vector corresponding to the first cycle is generated; the second data output vector generated in the second cycle is concatenated with the first data input vector to obtain the total data input vector corresponding to the first cycle; each of the multiple activation functions included in the pre-built temperature prediction model is used as the target activation function, and the total data input vector corresponding to the first cycle, the target weight matrix corresponding to the target activation function, and the target bias vector are input into the target activation function to obtain the transformation vector output by the target activation function; a first state vector corresponding to the second cycle is obtained; based on the transformation vector output by each activation function and the first state vector, a... The first data output vector of the first cycle is described; after generating the data output vector corresponding to the last cycle among the multiple cycles, the data output vector corresponding to the last cycle is dimensionally transformed according to the first weight matrix among the multiple weight matrices included in the temperature prediction model to obtain the predicted temperature vector. The first cycle is any cycle other than the first cycle among the multiple cycles, and the second cycle is the previous cycle of the first cycle. The predicted temperature vector includes the predicted temperature value of each heating component in the next cycle corresponding to the current cycle. The weight matrix is ​​used to perform linear transformation on the input vector, the activation function is used to perform nonlinear transformation on the input vector, and the bias vector is used to adjust the activation degree of the corresponding activation function. Based on the actual temperature value of the target heating component in the current cycle and the predicted temperature value in the next cycle, it is determined whether to perform a heat dissipation operation on the target heating component, wherein the target heating component is any one of at least one of the heating components; When it is determined that a heat dissipation operation will be performed on the target heat-generating component, the actual temperature value of the target heat-generating component in the current cycle is input into the pre-built heat dissipation control model to obtain the target control quantity; Based on the target control quantity, the heat dissipation operation quantity of the target heat dissipation component within the preset distance range of the target heat-generating component is increased.

2. The method according to claim 1, characterized in that, The step of generating the first data output vector for the first period based on the transformation vector output by each activation function and the first state vector includes: Based on the transformation vectors output by the activation functions other than the preset activation function and the first state vector, determine the second state vector corresponding to the first period; The first data output vector is determined based on the transformation vector output by the preset activation function and the second state vector.

3. The method according to claim 1 or 2, characterized in that, Each activation function is used as a target activation function. The total data input vector corresponding to the first period, the target weight matrix corresponding to the target activation function, and the target bias vector are input into the target activation function to obtain the transformation vector output by the target activation function, using the following expression: in, Let J be the transformation vector output by the target activation function, where J is the target activation function, which can be a sigmoid function or a hyperbolic tangent function. This is the total data input vector for the first period. The first data input vector for the first period. This is the second data output vector for the second period. Let be the target bias vector. Let be the target weight matrix.

4. The method according to claim 2, characterized in that, The second state vector corresponding to the first period is determined based on the transformation vectors output by the activation functions other than the preset activation function and the first state vector, using the following expression: in, This is the second state vector. The transformation vector output by the first activation function among the plurality of activation functions. This is the first state vector. The transformation vector is the output of the second activation function among the plurality of activation functions. The transformation vector is the output of the third activation function among the multiple activation functions.

5. The method according to claim 2 or 4, characterized in that, The first data output vector is determined based on the transformation vector output by the preset activation function and the second state vector, using the following expression: in, Output vector for the first data. The transformation vector output by the preset activation function. This is the second state vector.

6. The method according to any one of claims 1, 2, and 4, characterized in that, The step of determining whether to perform a heat dissipation operation on the target heat-generating component based on the actual temperature value of the target heat-generating component in the current cycle and the predicted temperature value in the next cycle includes: The temperature change value of the target heating element is obtained by subtracting the predicted temperature value of the target heating element in the next cycle from the actual temperature value of the target heating element in the current cycle. Determine whether the temperature change value of the target heating element is greater than or equal to a preset temperature change threshold; When it is determined that the temperature change value of the target heating component is greater than or equal to the temperature change threshold, it is determined that a heat dissipation operation will be performed on the target heating component. or, When it is determined that the temperature change value of the target heat-generating component is less than the temperature change threshold, it is determined that no heat dissipation operation will be performed on the target heat-generating component.

7. An apparatus for performing a heat dissipation operation, characterized in that, The device is applied to a server, the server including at least one heat-generating component, and the device includes: The acquisition module is used to acquire the actual temperature value of each of the heating components in multiple cycles during the current cycle, wherein the multiple cycles include the current cycle and at least one cycle before the current cycle; The determination module is used to generate a first data input vector corresponding to the first cycle based on the actual temperature value of each heating component in the first cycle; concatenate the second data output vector generated in the second cycle with the first data input vector to obtain the total data input vector corresponding to the first cycle; take each of the multiple activation functions included in the pre-built temperature prediction model as the target activation function, and input the total data input vector corresponding to the first cycle, the target weight matrix, and the target bias vector corresponding to the target activation function into the target activation function to obtain the transformation vector output by the target activation function; obtain the first state vector corresponding to the second cycle; generate the first data output vector of the first cycle based on the transformation vector output by each activation function and the first state vector; and generate the data output vector corresponding to the last cycle among the multiple cycles. After measurement, based on the first weight matrix among the multiple weight matrices included in the temperature prediction model, the data output vector corresponding to the last cycle is transformed to obtain a predicted temperature vector. The first cycle is any cycle other than the first cycle among the multiple cycles, and the second cycle is the cycle preceding the first cycle. The predicted temperature vector includes the predicted temperature value of each heating component in the next cycle corresponding to the current cycle. The weight matrix is ​​used to perform a linear transformation on the input vector, the activation function is used to perform a nonlinear transformation on the input vector, and the bias vector is used to adjust the activation degree of the corresponding activation function. Based on the actual temperature value of the target heating component in the current cycle and the predicted temperature value in the next cycle, it is determined whether to perform a heat dissipation operation on the target heating component, wherein the target heating component is any one of at least one of the heating components. An enhancement module is used to, when it is determined that a heat dissipation operation will be performed on the target heat-generating component, input the actual temperature value of the target heat-generating component in the current cycle into a pre-built heat dissipation control model to obtain a target control quantity; and, based on the target control quantity, enhance the heat dissipation operation quantity of the target heat-generating component within a preset distance range of the target heat-generating component.

8. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of performing heat dissipation operation as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method of performing heat dissipation operation as described in any one of claims 1 to 6.

Images