Fan control method and device, product, server and medium

Abstract

The invention discloses a fan control method and device, a product, a server and a medium, and relates to the field of computer hardware control. Under the condition that the temperature value of the server component is larger than or equal to the new current temperature threshold value, the target rotating speed value of the fan is output through the proportional-integral-differential control model, then the fan is controlled to rotate according to the target rotating speed value, and cooling of the server component through the fan is achieved; secondly, when the temperature threshold value is determined, the historical temperature values of the components under the multiple environment temperatures are considered, and the temperature threshold value is updated, that is, the fan is controlled by depending on the dynamic threshold value. The threshold value is adjusted in advance according to the environment temperature, so that frequent triggering of starting of the fan due to transient fluctuation of the environment temperature is avoided, and the stability of the server is ensured; and thirdly, based on an environment-load combined compensation mechanism, the rotating speed of the fan is dynamically adjusted to offset the influence of the environment temperature, the influence of the environment temperature and load factors on fan control is avoided as much as possible, and the fan control precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer hardware control, and in particular to a fan control method, device, product, server and medium. Background Art

[0002] Components in a server, such as the Central Processing Unit (CPU), generate heat during operation. To prevent performance degradation or damage due to overheating, fans are typically turned on in the server to cool the components and ensure server stability.

[0003] In the fan control process, related technologies rely on the maximum temperature limit (i.e., static threshold) specified in the product manual to control fan start and stop. When the component temperature is detected to be greater than or equal to the static threshold, the fan is activated to reduce the component temperature; otherwise, the fan is not activated. However, in practice, if the ambient temperature of the server is high, after using the fan to cool the component, the temperature will quickly rise again due to the influence of the ambient temperature. When the temperature exceeds or equals the static threshold, the fan is activated again to reduce the temperature. In other words, if the ambient temperature of the server is high, relying on the static threshold set in the product manual for fan control will cause the fan to be triggered frequently, resulting in increased wear of components (fans and components), affecting the stability and service life of the server.

[0004] Therefore, how to avoid frequent fan triggering as much as possible to ensure the stability and service life of the server is a technical problem that people in this field need to solve urgently. Summary of the Invention

[0005] The purpose of the present invention is to provide a fan control method, device, product, server, and medium to address the problem of relying on static thresholds set in the product manual for fan control, which frequently triggers the fan and causes increased wear of components (such as fans and parts), affecting the stability and service life of the server.

[0006] To solve the above technical problems, the present invention provides a fan control method, comprising: Obtaining a current temperature threshold for indicating starting fan control, and obtaining historical temperature values ​​of server components under multiple ambient temperatures; Updating the current temperature threshold using the historical temperature value to obtain a new current temperature threshold; Obtaining a current temperature value of the server component; When it is detected that the current temperature value is greater than or equal to the new current temperature threshold, obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; The target speed value of the fan is outputted through the proportional-integral-differential control model, and the fan is controlled to rotate according to the target speed value.

[0007] On the one hand, updating the current temperature threshold using the historical temperature value to obtain a new current temperature threshold includes: Get the peak temperature value in the historical temperature values; Performing weighted averaging processing on the temperature peak values ​​and obtaining a result after the weighted averaging processing; The new current temperature threshold is determined according to the current temperature threshold and the result after the weighted average processing.

[0008] On the other hand, obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server includes: Obtaining control parameters of a proportional-integral-derivative control model, and obtaining a first temperature difference between the current temperature value and the new current temperature threshold; Establishing a proportional term, an integral term, and a differential term according to the control parameter and the first temperature difference; Determining a compensation coefficient corresponding to an item representing environmental compensation according to the ambient temperature value and the load condition of the server, and obtaining a second temperature difference between the current ambient temperature value and the reference ambient temperature value; establishing a term for characterizing environmental compensation according to the compensation coefficient and the second temperature difference; The proportional-integral-derivative control model is determined according to the proportional term, the integral term, the differential term, and the term for characterizing environmental compensation.

[0009] On the other hand, the compensation coefficients corresponding to the items used to characterize environmental compensation are determined according to the ambient temperature value and the load condition of the server, including: Get the load difference between the current load of the server and the maximum load supported by the server; Get the ratio between the load difference and the maximum load supported by the server; Obtaining a product of the rate of change of the ambient temperature and the ratio; The compensation coefficient corresponding to the item characterizing the environmental compensation is determined according to the preset reference compensation coefficient and the multiplication result.

[0010] On the other hand, the fan control mode includes at least a manual control mode, an automatic control mode through a proportional-integral-differential control model, or a control mode determined according to power consumption; wherein the manual control mode has the highest priority among all fan control modes.

[0011] On the other hand, after detecting that the current temperature value is greater than or equal to the new current temperature threshold, and before obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server, the method further includes: Get the current fan control mode; When it is detected that the current fan control mode is the automatic control mode using the proportional-integral-differential control model, entering the step of obtaining the proportional-integral-differential control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; When it is detected that the current fan control mode is the manual control mode, a rotation speed value input by a user is obtained; the rotation speed value input by the user is used as a target rotation speed value of the fan, and the fan is controlled to rotate according to the target rotation speed value.

[0012] On the other hand, the fan control method also includes: When it is detected that the current temperature value is less than the new current temperature threshold, determining the fan control mode to be a control mode determined according to power consumption; In the control mode determined according to the power consumption, the rotation speed value of the fan is controlled to be a preset rotation speed value; wherein the preset rotation speed value is smaller than the target rotation speed value.

[0013] On the other hand, before detecting that the current temperature value is greater than or equal to the new current temperature threshold, the method further includes: Obtaining a first temperature difference between the current temperature value and the new current temperature threshold to determine a magnitude relationship between the current temperature value and the new current temperature threshold; Before obtaining a first temperature difference between the current temperature value and the new current temperature threshold, the method further includes: If a speed value input by the user is detected, determining that the current fan control mode is the manual control mode; using the speed value input by the user as the target speed value of the fan, and controlling the fan to rotate according to the target speed value; If the rotation speed value input by the user is not detected, the process proceeds to the step of obtaining the first temperature difference between the current temperature value and the new current temperature threshold.

[0014] On the other hand, after controlling the rotation speed of the fan to be a preset rotation speed value, the method further includes: Acquire the ambient temperature value at each moment within a preset time period and the temperature value of the server component under the ambient temperature value to obtain the historical temperature value, and return to the step of updating the current temperature threshold using the historical temperature value.

[0015] On the other hand, before obtaining the current temperature value of the server component, the method further includes: Obtain the historical temperature threshold corresponding to the speed manually set by the user and the frequency of manual user intervention; Determine the threshold weight based on the frequency of manual user intervention; A temperature threshold is determined according to the historical temperature threshold and the threshold weight to serve as the new current temperature threshold.

[0016] On the other hand, before detecting the magnitude relationship between the current temperature value and the new current temperature threshold, the method further includes: determining a new current temperature threshold to be used; The determination of the new current temperature threshold to be used includes: A new current temperature threshold value obtained after the historical temperature value is updated to the current temperature threshold value within a preset time since the server starts processing the service is used as the new current temperature threshold value to be used; a temperature threshold value determined based on the historical temperature threshold value and the threshold weight is used as the new current temperature threshold value to be used after the preset time since the server starts processing the service; Or, get the current ambient temperature; Obtain the historical temperature conditions when the historical temperature value updates the current temperature threshold; When it is detected that the current ambient temperature is the same as the historical temperature, the new current temperature threshold obtained by updating the historical temperature value to the current temperature threshold is used as the new current temperature threshold; Alternatively, a new current temperature threshold value obtained by updating the current temperature threshold value with the historical temperature value is used as the initial temperature threshold value; After determining a temperature threshold according to the historical temperature threshold and the threshold weight, the initial temperature threshold is replaced by the temperature threshold determined according to the historical temperature threshold and the threshold weight to serve as a new current temperature threshold used.

[0017] On the other hand, obtaining the current temperature value of the server component includes: Inputting historical temperature values ​​of the server components into a pre-established time series-based prediction model; The current temperature value of the server component is outputted through the time series-based prediction model.

[0018] On the other hand, after the target speed value of the fan is outputted through the proportional-integral-differential control model, and before the fan is controlled to rotate according to the target speed value, the method further includes: Determining a temperature jump frequency of the server component according to a temperature value of the server component predicted by the time series-based prediction model; Obtaining a jitter attenuation coefficient; wherein the jitter attenuation coefficient is determined by a temperature change trend; Determine a final target speed value according to the target speed value, the temperature jump frequency and the jitter attenuation coefficient; The controlling the fan to rotate according to the target speed value includes: Control the fan to rotate according to the final target speed value.

[0019] On the other hand, after outputting the target speed value of the fan through the proportional-integral-differential control model and controlling the fan to rotate according to the target speed value, the method further includes: Recording the ambient temperature value, the temperature value of the server component, the temperature threshold, the control parameter, and the compensation coefficient in a log; Copy the logs in the master control module to the backup control module; The method further comprises: When a failure of the main control module is detected, the temperature threshold data and the control parameters are restored according to the historical log data recorded in the backup control module; A simplified proportional-integral-differential control model is used to output a target speed value of the fan; wherein the simplified proportional-integral-differential control model includes a proportional term and an integral term; The fan is controlled to rotate according to the target speed value, and prompt information for indicating that the main control module is abnormal is output.

[0020] On the other hand, the fan control method also includes: Acquire a first type of server whose load is greater than or equal to a preset value, and acquire a second type of server whose load is less than the preset value; controlling the fan deployed on the second type of server to rotate at a speed lower than that of the fan deployed on the first type of server; Alternatively, a connection is established between the fan deployed on the second type of server and the first type of server, and the fan deployed on the second type of server is controlled to rotate to reduce the temperature of the server components on the first type of server.

[0021] On the other hand, after detecting that the current temperature value is greater than or equal to the new current temperature threshold, and before obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server, the method further includes: Comparing the relationship between the current temperature value and the preset temperature value; wherein the preset temperature value is greater than the new current temperature threshold; If the comparison result is that the current temperature value is less than or equal to the preset temperature value, then entering the step of obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; If the comparison result is that the current temperature value is greater than the preset temperature value, after regulating the temperature of the server component through the liquid cooling mode, the process returns to the step of comparing the relationship between the current temperature value and the preset temperature value.

[0022] In order to solve the above technical problems, the present invention further provides a fan control device, comprising: A first acquisition module is used to acquire a current temperature threshold for indicating starting fan control, and to acquire historical temperature values ​​of server components under multiple ambient temperatures; An updating module, configured to update a current temperature threshold using the historical temperature value to obtain a new current temperature threshold; A second acquisition module is used to obtain the current temperature value of the server component; a third acquisition module, configured to, when detecting that the current temperature value is greater than or equal to the new current temperature threshold, acquire a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; The output and control module is used to output the target speed value of the fan through the proportional-integral-differential control model and control the fan to rotate according to the target speed value.

[0023] In order to solve the above technical problem, the present invention further provides a computer program product, including a computer program / instruction, which implements the steps of the above fan control method when executed by a processor.

[0024] In order to solve the above technical problems, the present invention further provides a server, comprising: memory for storing computer programs; A processor is configured to implement the steps of the above-mentioned fan control method when executing the computer program.

[0025] In order to solve the above technical problem, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above fan control method are implemented.

[0026] The beneficial effects of the present invention are that, first, in this method, when the temperature value of the server component is greater than or equal to the new current temperature threshold, the proportional-integral-differential control model is used to output the target speed value of the fan, and then the fan is controlled to rotate according to the target speed value, thereby realizing the cooling of the server components by the fan; secondly, compared with the previous method of controlling the fan by relying on the static threshold of the component calibrated in the product manual, the method provided by the present invention, when determining the temperature threshold, takes into account the historical temperature values ​​of the component under multiple ambient temperatures and updates the temperature threshold, that is, relies on the dynamic threshold to control the fan. Since the threshold is adjusted in advance according to the ambient temperature, frequent triggering of fan startup due to short-term fluctuations in ambient temperature is avoided, thereby ensuring the stability and service life of the server; secondly, based on the proportional-integral-differential control model determined by considering the ambient temperature and load conditions, the target speed value of the fan is output, and the fan is controlled to rotate according to the target speed value. By quantitatively analyzing the fan speed value, precise control of the fan is achieved. Moreover, when determining the target speed value of the fan, the ambient temperature and load conditions are taken into account, that is, based on the environment-load joint compensation mechanism, the fan speed is dynamically adjusted to offset the influence of the ambient temperature, thereby avoiding the influence of ambient temperature and load factors on fan control as much as possible, and further improving the accuracy of fan control.

[0027] Furthermore, when determining the new current temperature threshold, a weighted average is performed on the temperature peaks in the historical temperature values, and the result of this weighted average is used to update the current temperature threshold. Because the temperature threshold determined based on the temperature peaks in the historical temperature values ​​is higher than the temperature threshold determined based on non-temperature peaks in the historical temperature values, fan control based on the temperature threshold determined by this method can further reduce the frequency of fan control triggered by server component temperatures, thereby ensuring server stability and service life.

[0028] The proportional, integral, and differential terms of the PID control model are determined based on the control parameters of the model and a first temperature difference (i.e., the difference between the current temperature of the server component and a new current temperature threshold). The term representing environmental compensation is determined based on a compensation coefficient and a second temperature difference (i.e., the difference between the current ambient temperature of the server component and a reference ambient temperature). The compensation coefficient is determined based on the ambient temperature and the server load. This method accurately establishes the PID control model.

[0029] Fan control modes include at least manual control, automatic control using a proportional-integral-derivative (PI-D) control model, and a control mode determined by power consumption. These multiple control modes provide flexible fan control. Furthermore, manual control has the highest priority among all fan control modes, meeting user fan control needs.

[0030] After detecting that the current temperature value is greater than or equal to the new current temperature threshold, in addition to controlling the fan through the automatic control mode of the proportional-integral-differential control model, the fan can also be controlled through the manual control mode, realizing flexible control of the fan and achieving the effect of cooling the server components.

[0031] When it is detected that the current temperature value is less than the new current temperature threshold, the fan is controlled using a control mode determined according to the power consumption (ie, a low power consumption mode), thereby reducing the power consumption of the server.

[0032] During fan control, the fan is first manually controlled. Based on the relationship between the current temperature and the new current temperature threshold, the fan is then automatically controlled using a proportional-integral-derivative control model, or a control mode determined by power consumption. This provides relatively flexible fan control.

[0033] After controlling the fan according to the control mode determined by the power consumption, the ambient temperature value and the temperature value of the server component under the ambient temperature value are collected to obtain the historical temperature value, and the step of using the historical temperature value to update the current temperature threshold is continued to return to achieve the update of the temperature threshold and the continuous control of the fan, thereby achieving continuous control of the temperature of the server component.

[0034] The threshold weight is determined based on the frequency of manual user intervention. The temperature threshold is determined based on the historical temperature threshold corresponding to the user's manually set speed and the threshold weight, and used as the new current temperature threshold. In other words, based on manual intervention records, the system can automatically optimize the temperature threshold, reducing the need for frequent user intervention.

[0035] Before detecting the magnitude relationship between the current temperature value and the new current temperature threshold, the new current temperature threshold to be used can be determined from the new current temperature threshold obtained after updating the current temperature threshold with the historical temperature value, as well as the temperature threshold determined based on the historical temperature threshold and the threshold weight. This improves the flexibility of obtaining the temperature threshold.

[0036] The current temperature value of server components is output by a time series-based prediction model. Compared with the method of measuring the actual temperature value through sensors, it can control the fan in advance, thereby achieving preemptive cooling of server components.

[0037] After the fan's target speed is output through the proportional-integral-derivative control model, the final target speed is determined based on the target speed, the temperature jump frequency, and the jitter attenuation coefficient. The fan is then controlled to rotate at the final target speed. This method reduces the fan speed fluctuation amplitude based on the jitter attenuation coefficient.

[0038] If a failure of the main control module is detected, temperature threshold data and control parameters are restored based on historical log data recorded in the backup control module. A simplified proportional-integral-derivative control model, including proportional and integral terms, is used to output the fan's target speed. The fan is then controlled to rotate at the target speed and a prompt message indicating a main control module failure is output. This method ensures fan control through redundant control, thereby achieving temperature regulation of server components.

[0039] When the server load is less than the preset value, the fan speed on the server is controlled to decrease, reducing the server's power consumption; or, the fan resources on the server are allocated to servers with a load greater than or equal to the preset value, realizing dynamic allocation of cooling resources.

[0040] If the current temperature of a server component is determined to be greater than or equal to a new current temperature threshold and greater than a preset temperature value (where the preset temperature value is greater than the new current temperature threshold), the server component is first cooled using liquid cooling. Subsequently, if the current temperature of the server component is determined to be less than the new current temperature threshold, the fan is used to cool the server component. This method improves the efficiency of cooling the server component.

[0041] In addition, the present invention also provides a fan control device, a computer program product, a server, and a computer-readable storage medium, which have the same or corresponding technical features as the above-mentioned fan control method and have the same effects as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0043] Figure 1 A schematic diagram of a system for controlling a fan provided by an embodiment of the present invention; Figure 2 A flow chart of a fan control method provided by an embodiment of the present invention; Figure 3A flowchart of a method for controlling a fan based on a proportional-integral-derivative control model provided by an embodiment of the present invention; Figure 4 A flowchart of a multi-mode fan control method provided by an embodiment of the present invention; Figure 5 A structural diagram of a server provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0045] The core of the present invention is to provide a fan control method, device, product, server and medium to solve the problem of relying on static thresholds set in the product manual for fan control, which will frequently trigger the fan and cause increased wear of components (such as fans and parts), affecting the stability and service life of the server.

[0046] In order to enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods. Figure 1 A schematic diagram of a system for controlling a fan provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the system includes an environmental sensing module, a dynamic threshold correction module, and a multi-mode control module. The environmental sensing module collects temperature data of server components and ambient temperature data. The dynamic threshold correction module updates the set temperature threshold based on the ambient temperature data collected by the environmental sensing module. The multi-mode control module includes multiple fan control modes, such as manual control mode, automatic control mode using a proportional-integral-derivative control model, or a control mode determined by power consumption (i.e., low-power mode). The multi-mode control module is used to control the fan to operate according to the control mode to achieve temperature regulation of server components.

[0047] Figure 2 A flow chart of a fan control method provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown, the method includes: S10: Obtaining a current temperature threshold for indicating starting fan control, and obtaining historical temperature values ​​of server components under multiple ambient temperatures; S11: Update the current temperature threshold using the historical temperature value to obtain a new current temperature threshold; S12: Obtain the current temperature value of the server component; S13: When it is detected that the current temperature value is greater than or equal to the new current temperature threshold, a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server is obtained; S14: Outputting a target speed value of the fan through a proportional-integral-differential control model, and controlling the fan to rotate according to the target speed value.

[0048] Server components such as CPU, etc. Use temperature sensors to collect CPU temperature in real time ( ) and ambient temperature ( ), the data sampling frequency is, for example, 1 Hz. The ambient temperature sensor is deployed at the device's air inlet. After obtaining the temperature value, the temperature data can be integerized (for example, by removing the decimal point). The data is recorded in real time to the log file / home / icfs / info_CPU_log for subsequent analysis.

[0049] The current temperature threshold obtained for initiating fan control is not limited. To achieve a reasonable setting of the temperature threshold and prevent frequent triggering of the server component temperature, the present invention updates the current temperature threshold based on historical temperature values ​​of the server component at multiple ambient temperatures, thereby obtaining a new current temperature threshold. The historical temperature values ​​selected are not limited. However, to prevent frequent triggering of fan control by server component temperature, the new current temperature threshold is determined based on the peak temperature value in the historical temperature values.

[0050] Specifically, updating the current temperature threshold using the historical temperature value to obtain a new current temperature threshold includes: Get the peak temperature value in the historical temperature values; Performing weighted averaging processing on the temperature peak values ​​and obtaining the weighted averaging processing result; A new current temperature threshold is determined according to the current temperature threshold and the result of weighted average processing.

[0051] The dynamic threshold correction module is based on the historical peak temperature ( ), update the temperature threshold ( ). The formula for updating the temperature threshold is: ; in, Indicates the current temperature threshold. Indicates the new current temperature threshold, is the historical data weight coefficient, , determine the optimal value through experiments (default 0.2); Indicates the total number of temperature peaks, and i indicates the number of the current temperature peak.

[0052] In implementation, in order to obtain the new current temperature threshold, the historical peak temperature (such as the highest temperature within 24 hours) can be extracted from the log file; the mean deviation is calculated and the weight coefficient is used. (Default 0.2) Dynamically adjust the threshold. For example: if the historical peak values ​​are 80°C, 82°C, and 78°C, and the current temperature threshold (i.e., the original temperature threshold) is 75°C, then the new current temperature threshold (i.e., the new threshold) is: .

[0053] In the above description, the temperature threshold is determined based on the historical temperature threshold. In practice, the temperature threshold can also be determined by learning user behavior. Specifically, before obtaining the current temperature value of the server component, the following steps are also included: Obtain the historical temperature threshold corresponding to the speed manually set by the user and the frequency of manual user intervention; Determine the threshold weight based on the frequency of manual user intervention; A temperature threshold is determined according to the historical temperature threshold and the threshold weight to serve as a new current temperature threshold.

[0054] During implementation, the frequency of manual intervention is recorded and the initial value of the temperature threshold is dynamically adjusted. , its learning formula is: ; in, Indicates the historical temperature peak value corresponding to the speed manually set by the user; is the default threshold weight.

[0055] In this method, based on the manual intervention records, the system can automatically optimize the temperature threshold, reducing the need for frequent user intervention.

[0056] In order to select an appropriate temperature threshold, before detecting the relationship between the current temperature value and the new current temperature threshold, the following steps are also performed: Determines the new current temperature threshold to use.

[0057] The determination of the new current temperature threshold to be used includes: Method 1: From the start of the server processing business, the new current temperature threshold obtained by updating the current temperature threshold with the historical temperature value within a preset time is used as the new current temperature threshold; from the start of the server processing business, after a preset time, the temperature threshold determined based on the historical temperature threshold and the threshold weight is used as the new current temperature threshold.

[0058] Method 2: Get the current ambient temperature; Get the historical temperature conditions when using historical temperature values ​​to update the current temperature threshold; When it is detected that the current ambient temperature condition is the same as the historical temperature condition, a new current temperature threshold value obtained by updating the current temperature threshold value with the historical temperature value is used as the new current temperature threshold value.

[0059] Method 3: The new current temperature threshold obtained by updating the historical temperature value is used as the initial temperature threshold; After the temperature threshold is determined according to the historical temperature threshold and the threshold weight, the initial temperature threshold is replaced by the temperature threshold determined according to the historical temperature threshold and the threshold weight to serve as the new current temperature threshold used.

[0060] In this method, before detecting the magnitude relationship between the current temperature value and the new current temperature threshold, the new current temperature threshold to be used can be determined from the new current temperature threshold obtained by updating the current temperature threshold with the historical temperature value, and the temperature threshold determined based on the historical temperature threshold and the threshold weight. This improves the flexibility of obtaining the temperature threshold.

[0061] If the server component's current temperature is detected to be greater than or equal to a new current temperature threshold, a proportional-integral-derivative (PID) control model is obtained, which is determined by the new current temperature threshold, the current temperature, the ambient temperature, and the server's load. The PID model outputs the fan's target speed (i.e., the target speed).

[0062] Obtaining a proportional-integral-derivative control model determined by a new current temperature threshold, a current temperature value, an ambient temperature value, and a load condition of the server includes: Obtaining control parameters of a proportional-integral-derivative control model, and obtaining a first temperature difference between a current temperature value and a new current temperature threshold; Establishing a proportional term, an integral term, and a differential term according to the control parameter and the first temperature difference; Determining a compensation coefficient corresponding to an item representing environmental compensation according to the ambient temperature value and the load condition of the server, and obtaining a second temperature difference between the current ambient temperature value and the reference ambient temperature value; Establishing a term for characterizing environmental compensation according to a compensation coefficient and a second temperature difference; A proportional-integral-derivative control model is determined according to a proportional term, an integral term, a differential term, and a term for characterizing environmental compensation (ie, an environmental compensation term).

[0063] The compensation coefficient corresponding to the item representing the environmental compensation is determined according to the ambient temperature value and the load condition of the server, including: Get the load difference between the current load of the server and the maximum load supported by the server; Get the ratio between the load difference and the maximum load supported by the server; Get the product of the rate of change of the ambient temperature and the ratio; The compensation coefficient corresponding to the item characterizing the environmental compensation is determined according to the preset benchmark compensation coefficient and the product result.

[0064] Specifically, the control parameters of the proportional-integral-derivative control model include the weight corresponding to the proportional term ( ), the weight corresponding to the integral term ( ) and the weight corresponding to the differential term ( In practice, PID control parameters can be optimized in real time using machine learning algorithms (such as deep reinforcement learning). Based on historical temperature data and workload patterns, the system automatically adjusts parameters to address different scenarios (such as sudden high loads or long periods of low loads), further improving control accuracy and energy efficiency.

[0065] The compensation coefficient is expressed as (The default value is 0.3). The expression of the PID algorithm provided by the present invention (i.e., the improved PID algorithm with environmental compensation) is: ; in, Indicates the fan speed value (i.e. target speed value); Indicates the reference ambient temperature value (i.e., the reference ambient temperature), which is usually set to 25°C; Indicates the first temperature difference between the current temperature value and the new current temperature threshold. If the server component is a CPU, then ; is the proportional term, is the integral term, is the differential term; It is an environmental compensation item.

[0066] Compensation coefficient The expression is: ; in, represents the reference compensation coefficient, is the weight factor, is the difference between the current load and the maximum load; The maximum load capacity of the system.

[0067] For example, if the ambient temperature rises by 5°C, the compensation term automatically increases the speed by 10%. This compensation term can offset the negative impact of high temperatures on heat dissipation efficiency. In the PID control algorithm, the ambient temperature compensation coefficient γ is dynamically adjusted based on the real-time ambient temperature change rate and load fluctuations. This method dynamically adjusts the compensation strength through dual feedback from load and ambient temperature, improving temperature control accuracy in complex scenarios. Experimental data shows that dynamically adjusting the compensation factor can reduce temperature fluctuations by 15%-20%, making it particularly suitable for high-load servers and industrial equipment.

[0068] Figure 3 A flow chart of a method for controlling a fan based on a proportional-integral-differential control model provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, the method includes: S15: Temperature collection preprocessing.

[0069] Specifically, obtain the temperature value of the central processing unit in real time And calculate the temperature deviation .

[0070] S16: The proportional-integral-derivative control algorithm calculates the proportional term, the integral term, and the differential term.

[0071] S17: Ambient temperature compensation, calculate the ambient compensation item.

[0072] S18: Compensation superposition, adjust the speed according to the ambient temperature.

[0073] S19: Calculate the total fan speed.

[0074] Specifically, .

[0075] S20: Execute the fan speed command and issue a fan speed instruction.

[0076] After the target speed value of the fan is output through the PID model, the fan is controlled to rotate according to the target speed value to achieve cooling of the server components.

[0077] In this method, when the temperature of a server component is greater than or equal to a new current temperature threshold, a proportional-integral-derivative control model is used to output a target fan speed value. The fan is then controlled to rotate at the target speed, achieving cooling of server components using the fan. Secondly, compared to previous fan control methods that rely on static component thresholds specified in product manuals, this method considers historical component temperature values ​​at multiple ambient temperatures when determining the temperature threshold and updates the temperature threshold, thus relying on a dynamic threshold for fan control. Because the threshold is pre-adjusted based on the ambient temperature, frequent fan activation caused by brief ambient temperature fluctuations is avoided, ensuring server stability and service life. Thirdly, a proportional-integral-derivative control model, determined based on ambient temperature and load conditions, outputs a target fan speed value, controlling the fan to rotate at the target speed value. This quantitative analysis of the fan speed values ​​enables precise fan control. Furthermore, the ambient temperature and load conditions are considered when determining the fan target speed value. Based on a combined ambient-load compensation mechanism, the fan speed is dynamically adjusted to offset the effects of ambient temperature, minimizing the impact of ambient temperature and load factors on fan control and further improving fan control accuracy.

[0078] In related fan control, only a single control mode is usually used to control the fan, which makes it impossible to flexibly control the fan and, in turn, to cool the server components. For example, only PID automatic adjustment is supported (considering only the proportional, integral, and differential terms), which lacks the ability for emergency manual intervention and makes it difficult to cope with sudden loads (such as a surge in instantaneous computing demand in a data center). Therefore, in the method provided by the present invention, the fan control mode includes at least a manual control mode, an automatic control mode using a proportional-integral-differential control model, or a control mode determined based on power consumption (low-power mode); wherein, the manual control mode has the highest priority among all fan control modes.

[0079] The automatic control mode of the proportional-integral-differential control model has been described in detail above and will not be repeated here. The manual control mode is: obtain the speed value input by the user; use the speed value input by the user as the target speed value of the fan, and control the fan to rotate according to the target speed value. If the user enters a hexadecimal speed value (such as 0x01 to 0xFF), the fan is directly controlled. This mode can override the automatic control and is used in emergency cooling scenarios. In low power mode, the speed value of the fan is controlled to a preset speed value; wherein the preset speed value is less than the target speed value. For example, when When the fan speed is reduced to the preset speed value (such as 0x20).

[0080] In this method, flexible control of the fan is achieved through multiple control modes; moreover, the priority of the manual control mode is the highest among all fan control modes, meeting the user's demand for fan control.

[0081] After detecting that the current temperature value is greater than or equal to a new current temperature threshold, and before obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server, the following further comprises: Get the current fan control mode; When it is detected that the current fan control mode is an automatic control mode using a proportional-integral-differential control model, entering a step of obtaining a proportional-integral-differential control model determined by a new current temperature threshold, a current temperature value, an ambient temperature value, and a load condition of the server; When it is detected that the current fan control mode is the manual control mode, a rotation speed value input by a user is obtained; the rotation speed value input by the user is used as a target rotation speed value of the fan, and the fan is controlled to rotate according to the target rotation speed value.

[0082] In this method, after detecting that the current temperature value is greater than or equal to the new current temperature threshold, in addition to being able to control the fan through the automatic control mode of the proportional-integral-differential control model, the fan can also be controlled through the manual control mode, thereby realizing flexible control of the fan and achieving the effect of cooling the server components.

[0083] In practice, the manual control mode has the problem of insufficient hardware compatibility. To improve hardware compatibility, the speed value entered by the user is used as the fan target speed value, including: Get the server type; Set the corresponding instruction containing the rotation speed value input by the user according to the server type; The speed value input by the user is used as the target speed value of the fan according to the instruction including the speed value input by the user.

[0084] Instructions such as Intelligent Platform Management Interface (IPMI) commands. Compared to general IPMI command control, which is not optimized for specific server architectures and can easily cause command conflicts or response delays, the method provided by the present invention customizes IPMI command sequences for different servers, optimizes command transmission timing, and reduces response delays (experimentally measured delays were reduced from 120ms to 70ms).

[0085] Furthermore, for small IoT devices (such as edge computing gateways), the IPMI instruction set and algorithm computational complexity have been optimized to achieve a low-power embedded version. For example, a lightweight PID algorithm (retaining only the proportional and integral terms) and simplified sensor data collection make it suitable for resource-constrained hardware environments.

[0086] In order to reduce the power consumption of the server, in implementation, the fan control method further includes: When it is detected that the current temperature value is less than the new current temperature threshold, determining the fan control mode to be a control mode determined according to power consumption; In the control mode determined according to the power consumption, the rotation speed value of the fan is controlled to be a preset rotation speed value; wherein the preset rotation speed value is smaller than the target rotation speed value.

[0087] In this method, the power consumption of the server is reduced by controlling the fan to rotate at a lower speed value.

[0088] In addition to the above description, when it is detected that the current temperature value is greater than or equal to the new current temperature threshold, the fan can be controlled in the automatic control mode of the proportional-integral-differential control model; or the fan can be controlled in the manual control mode; when it is detected that the current temperature value is less than the new current temperature threshold, the fan can be controlled in the low power consumption mode. This embodiment also provides another fan control scenario. Specifically, before detecting that the current temperature value is greater than or equal to the new current temperature threshold, it also includes: Obtaining a first temperature difference between the current temperature value and the new current temperature threshold to determine a magnitude relationship between the current temperature value and the new current temperature threshold; Before obtaining the first temperature difference between the current temperature value and the new current temperature threshold, the method further includes: If a speed value input by the user is detected, the current fan control mode is determined to be the manual control mode; the speed value input by the user is used as the target speed value of the fan, and the fan is controlled to rotate according to the target speed value; If the rotation speed value input by the user is not detected, the process proceeds to the step of obtaining a first temperature difference between the current temperature value and a new current temperature threshold.

[0089] In order to update the temperature threshold and continuously control the fan, after controlling the fan speed to the preset speed value, the following steps are also included: Acquire the ambient temperature values ​​at each moment within a preset time period and the temperature values ​​of the server components under the ambient temperature values ​​to obtain historical temperature values, and return to the step of updating the current temperature threshold using the historical temperature values.

[0090] There is no limit on the preset duration, which is determined based on actual conditions.

[0091] Figure 4A flowchart of a multi-mode fan control method provided by an embodiment of the present invention is shown in FIG. Figure 4 As shown, the method includes: S21: Determine whether a manual command is input; if so, proceed to step S22; if not, proceed to step S23; S22: Enter manual mode and execute the command to set the speed; S23: Calculating the temperature difference between the server component temperature and the new current temperature threshold; S24: Determine whether the temperature difference is greater than or equal to 0; if so, proceed to step S25; if not, proceed to step S26; S25: Enter the automatic control mode of the proportional-integral-differential control model and dynamically adjust the speed; S26: Enter low power mode and reduce the speed to a safe value; S27: Collect the temperature data of each sensor and the fan speed data within the set collection time; return to step S21; S28: Supplement the temperature model formula and incorporate the proportional-integral-differential control model into the automatic control mode regulation.

[0092] In implementation, the multi-mode fan control method includes: 1. System initialization, loading hardware adaptation parameters, and setting initial thresholds .

[0093] Real-time collection and ,according to Calculate the dynamic threshold and then Compare to dynamic threshold.

[0094] 3. Determine the control mode priority and perform speed adjustment.

[0095] Check whether there is manual instruction input, and execute it immediately if so; If there is no manual instruction, calculate the current temperature deviation ΔT. If ΔT ≥ 0, enable automatic PID mode; If ΔT<0, switch to low power mode.

[0096] 4. Record temperature, speed, and environmental data to the log for threshold correction and energy efficiency analysis.

[0097] If the temperature of the server components is measured by sensors and then adjusted, there will be a lag in the temperature control of the server components. Therefore, in order to cool the server components in advance, the current temperature value of the server components is obtained by: Inputting historical temperature values ​​of server components into a pre-established time series-based prediction model; Output the current temperature value of the server component through a time series-based prediction model.

[0098] This method uses a time series prediction algorithm to predict CPU temperature trends in advance and proactively adjust fan speeds to avoid excessive temperature fluctuations. For example, it can improve cooling capacity in advance of a data center's predicted computing workload peak.

[0099] To reduce fan speed fluctuations, after outputting the fan's target speed through the proportional-integral-derivative control model and before controlling the fan to rotate at the target speed, the following steps are also performed: Determining a temperature jump frequency of the server component according to a temperature value of the server component predicted by a time series-based prediction model; Obtaining a jitter attenuation coefficient; wherein the jitter attenuation coefficient is determined by a temperature change trend; The final target speed value is determined based on the target speed value, the temperature jump frequency and the jitter attenuation coefficient.

[0100] Controlling the fan to rotate according to the target speed value includes: Control the fan to rotate according to the final target speed value.

[0101] This method provides a temperature smoothing algorithm based on the jitter coefficient. The jitter coefficient table is introduced into the PID control, and the jump frequency ( ), dynamically adjust the fan speed adjustment range, the formula is: ; in, represents the final target speed value, and p represents the vibration attenuation coefficient.

[0102] This method dynamically optimizes the attenuation intensity by combining temperature change trends (such as differential terms) to reduce fan speed oscillation. In sudden high-load scenarios, this algorithm can reduce the fan speed fluctuation range from ±20% to ±5%, extending fan life. It is worth noting that in addition to server components such as the CPU, this method can also be applied to the heat dissipation control of high-performance computing chips such as graphics processing units (GPUs) and tensor processing units (TPUs). Because GPUs in artificial intelligence (AI) training and high-performance computing often face higher heat dissipation requirements, dynamic threshold correction and environmental compensation mechanisms can significantly improve energy efficiency.

[0103] To facilitate understanding of fan control, after outputting the fan's target speed value through the proportional-integral-derivative control model and controlling the fan to rotate at the target speed value, the following steps are also included: Record ambient temperature values, server component temperature values, temperature thresholds, control parameters, and compensation coefficients in a log; Copy the logs in the master control module to the backup control module.

[0104] Anomaly detection models are trained using temperature and speed log data to proactively identify fan aging or cooling system failures. For example, by analyzing deviations between speed and temperature response time, potential hardware issues can be alerted. When a fan failure is detected, the system automatically switches to a backup fan or adjusts other cooling modules (such as increasing liquid cooling power) to ensure continued stable system operation.

[0105] Fan control methods also include: When the main control module fails, the temperature threshold data and control parameters are restored according to the historical log data recorded in the backup control module; A simplified proportional-integral-differential control model is used to output a target speed value of the fan; wherein the simplified proportional-integral-differential control model includes a proportional term and an integral term; The fan is controlled to rotate according to the target speed value and a prompt message is output to indicate an abnormality of the main control module.

[0106] The system consists of a main control module (system CPU control circuit) and a backup monitoring module. If the main control module fails, the backup module takes over control through an independent communication circuit and performs the following operations: Restore the most recent valid threshold based on historical log data and compensation parameters; Enable the simplified PID algorithm (retain only the proportional and integral terms) to maintain basic temperature control function; Trigger alarm components (such as LEDs or buzzers) to indicate system abnormalities.

[0107] This method can be implemented based on a multi-level redundant control and fault self-healing module. In this module, combined with a user behavior learning module (such as manual intervention records), intelligent recovery based on thresholds and parameters can be achieved to avoid the limitations of traditional redundant systems that rely on fixed preset values.

[0108] In a server cluster or edge computing node, multiple devices collaborate to dissipate heat. To avoid local overheating or resource waste, fan control methods also include: Acquire a first type of server whose load is greater than or equal to a preset value, and acquire a second type of server whose load is less than a preset value; controlling the fans deployed on the second type of servers to rotate at a speed lower than that of the fans deployed on the first type of servers; Alternatively, a connection is established between a fan deployed on a second type of server and the first type of server, and the fan deployed on the second type of server is controlled to rotate to reduce the temperature of server components on the first type of server.

[0109] In this distributed thermal management approach, multiple devices collaborate to dissipate heat. By sharing ambient temperature data and load information, cooling resources are dynamically allocated to avoid local overheating or resource waste. For example, when a node has a low load, its fan speed can be reduced, and cooling resources can be allocated to the higher-loaded node.

[0110] To improve the efficiency of reducing the temperature of server components, after detecting that the current temperature value is greater than or equal to a new current temperature threshold, and before obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server, the following is further included: Comparing the current temperature value with the preset temperature value; wherein the preset temperature value is greater than the new current temperature threshold; If the comparison result is that the current temperature value is less than or equal to the preset temperature value, then entering the step of obtaining a proportional-integral-differential control model determined by a new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; If the comparison result shows that the current temperature value is greater than the preset temperature value, the temperature of the server component is regulated by the liquid cooling mode, and then the process returns to the step of comparing the relationship between the current temperature value and the preset temperature value.

[0111] This method combines liquid cooling technology to automatically switch between air and liquid cooling modes based on temperature thresholds. When the temperature exceeds the critical value, the liquid cooling system is activated to rapidly reduce the temperature. In practice, a PID algorithm can be used to coordinate the operating ratio of the two cooling methods.

[0112] The intelligent speed regulation system and method proposed in this paper, combining dynamic ambient temperature compensation, adaptive threshold correction, and multi-mode coordinated control, significantly improves the system's adaptability and control accuracy by introducing an ambient temperature compensation coefficient, a dynamic threshold correction algorithm, and multi-mode priority logic. It also achieves an optimal balance between heat dissipation efficiency and energy consumption, and is adaptable to heterogeneous hardware architectures. A user behavior learning module further optimizes the personalized adaptability of the control strategy. Applicable to scenarios such as servers and high-performance computing equipment, it achieves the dual goals of high-precision temperature control and optimized energy consumption.

[0113] In addition to the temperature control methods described above, dynamic fan speed adjustment can be used to reduce energy consumption during peak electricity demand periods, responding to the grid's load balancing needs. For example, by incorporating electricity price signals, cooling performance can be improved during low-price periods, while speed can be moderately reduced to save costs during peak periods. Furthermore, cooling strategies can be dynamically adjusted based on the real-time power supply from renewable energy sources (such as solar and wind power). For example, when solar power is abundant, a slightly higher temperature threshold can be allowed to reduce energy consumption; when power supply is limited, a stricter temperature control strategy can be implemented.

[0114] In the above embodiments, the fan control method is described in detail. The present invention also provides corresponding embodiments of a fan control device and a server. It should be noted that the present invention describes the embodiments of the device from two perspectives: one from the perspective of functional modules and the other from the perspective of hardware.

[0115] An embodiment of the present invention provides a fan control device, which includes: A first acquisition module is used to acquire a current temperature threshold for indicating starting fan control, and to acquire historical temperature values ​​of server components under multiple ambient temperatures; An updating module, configured to update the current temperature threshold using the historical temperature value to obtain a new current temperature threshold; A second acquisition module is used to obtain the current temperature value of the server component; A third acquisition module is configured to, when detecting that the current temperature value is greater than or equal to the new current temperature threshold, acquire a proportional-integral-differential control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; The output and control module is used to output the target speed value of the fan through a proportional-integral-differential control model and control the fan to rotate according to the target speed value.

[0116] In some embodiments, the update module includes: A fourth acquisition module is used to obtain a temperature peak value in the historical temperature values; The processing and acquisition module is used to perform weighted average processing on the temperature peak value and obtain the result after weighted average processing; The first determining module is configured to determine a new current temperature threshold according to the current temperature threshold and a result after weighted average processing.

[0117] In some embodiments, the third acquisition module specifically includes: a fifth acquisition module, configured to acquire control parameters of a proportional-integral-derivative control model and a first temperature difference between a current temperature value and a new current temperature threshold; A first establishing module, configured to establish a proportional term, an integral term, and a differential term according to a control parameter and a first temperature difference; A determination and acquisition module, configured to determine a compensation coefficient corresponding to an item characterizing environmental compensation according to an ambient temperature value and a load condition of a server, and to obtain a second temperature difference between a current ambient temperature value and a reference ambient temperature value; A second establishing module is used to establish a term for characterizing environmental compensation according to the compensation coefficient and the second temperature difference; The second determination module is used to determine a proportional-integral-derivative control model according to the proportional term, the integral term, the differential term, and the term used to characterize environmental compensation.

[0118] In some embodiments, the determination and acquisition module includes a third determination module for determining a compensation coefficient corresponding to an item representing environmental compensation according to an ambient temperature value and a load condition of the server.

[0119] The third determination module specifically includes: A first sub-acquisition module is used to obtain a load difference between the current load of the server and the maximum load supported by the server; The second sub-acquisition module is used to obtain the ratio between the load difference and the maximum load supported by the server; The third sub-acquisition module is used to obtain the product of the rate of change of the ambient temperature and the ratio; The fourth determining module is configured to determine a compensation coefficient corresponding to an item representing environmental compensation according to a preset benchmark compensation coefficient and a multiplication result.

[0120] In some embodiments, the fan control device further comprises: A sixth acquisition module, configured to acquire a current fan control mode; A first detection and triggering module is configured to trigger a third acquisition module when detecting that the current fan control mode is an automatic control mode using a proportional-integral-differential control model; The detection and acquisition module is used to obtain the speed value input by the user when it is detected that the current fan control mode is the manual control mode; use the speed value input by the user as the target speed value of the fan, and control the fan to rotate according to the target speed value.

[0121] In some embodiments, the fan control device further comprises: a first detection and determination module, configured to determine, when detecting that the current temperature value is less than a new current temperature threshold, that the fan control mode is a control mode determined according to power consumption; The first control module is configured to control the fan's rotation speed to a preset rotation speed value under a control mode determined according to power consumption; wherein the preset rotation speed value is smaller than a target rotation speed value.

[0122] In some embodiments, the fan control device further comprises: The seventh acquisition module is used to acquire a first temperature difference between the current temperature value and the new current temperature threshold value to determine a magnitude relationship between the current temperature value and the new current temperature threshold value.

[0123] The fan control also includes: The second detection and determination module is configured to determine that the current fan control mode is the manual control mode if a speed value input by the user is detected; use the speed value input by the user as the target speed value of the fan, and control the fan to rotate according to the target speed value; The second detection and triggering module is used to trigger the seventh acquisition module if no rotation speed value input by the user is detected.

[0124] In some embodiments, the fan control device further comprises: The eighth acquisition module is used to obtain the ambient temperature value at each moment within a preset time period and the temperature value of the server component under the ambient temperature value to obtain a historical temperature value and return it to the trigger update module.

[0125] In some embodiments, the fan control device further comprises: A fourth sub-acquisition module is used to obtain historical temperature thresholds corresponding to the rotation speed manually set by the user and the frequency of manual intervention by the user; a fifth determining module, configured to determine a threshold weight according to a frequency of manual intervention by a user; The sixth determining module is configured to determine a temperature threshold according to the historical temperature threshold and the threshold weight, so as to serve as a new current temperature threshold.

[0126] In some embodiments, the fan control device further comprises: a seventh determining module, configured to determine a new current temperature threshold to be used; The seventh determination module specifically includes: The first module is configured to use, as a new current temperature threshold value obtained after updating the current temperature threshold value with the historical temperature value within a preset time period since the server starts processing business, as the new current temperature threshold value used; and after a preset time period since the server starts processing business, use a temperature threshold value determined based on the historical temperature threshold value and the threshold weight as the new current temperature threshold value used; Or, a fifth sub-acquisition module is used to obtain the current ambient temperature; The sixth sub-acquisition module is used to obtain the historical temperature conditions when the historical temperature value updates the current temperature threshold; The second module is used to, when it is detected that the current ambient temperature is the same as the historical temperature, update the current temperature threshold with the historical temperature value to obtain a new current temperature threshold as the new current temperature threshold used; Alternatively, the third module is used to update the current temperature threshold with the historical temperature value, thereby obtaining a new current temperature threshold as the initial temperature threshold; The replacement module is used to replace the initial temperature threshold with the temperature threshold determined according to the historical temperature threshold and the threshold weight after determining the temperature threshold according to the historical temperature threshold and the threshold weight, so as to serve as the new current temperature threshold used.

[0127] In some embodiments, the second acquisition module includes: An input module, configured to input historical temperature values ​​of server components into a pre-established time series-based prediction model; The first output module is used to output the current temperature value of the server component through a time series-based prediction model.

[0128] In some embodiments, the fan control device further comprises: an eighth determining module, configured to determine a temperature jump frequency of the server component according to a temperature value of the server component predicted by a time series-based prediction model; a ninth acquisition module, configured to acquire a jitter attenuation coefficient, wherein the jitter attenuation coefficient is determined by a temperature variation trend; The ninth determination module is configured to determine a final target speed value according to the target speed value, the temperature jump frequency, and the jitter attenuation coefficient.

[0129] In some embodiments, the fan control device further comprises: A recording module, used to record ambient temperature values, temperature values ​​of server components, temperature thresholds, control parameters, and compensation coefficients into a log; The replication module is used to replicate the logs in the master control module to the backup control module.

[0130] The fan control also includes: A recovery module is used to restore temperature threshold data and control parameters based on historical log data recorded in the backup control module when a failure of the main control module is detected; A second output module is configured to output a target speed value of the fan using a simplified proportional-integral-differential control model, wherein the simplified proportional-integral-differential control model includes a proportional term and an integral term; The control and output module is used to control the fan to rotate according to the target speed value and output prompt information to indicate abnormality of the main control module.

[0131] In some embodiments, the fan control device further comprises: a tenth obtaining module, configured to obtain a first type of server having a load greater than or equal to a preset value, and obtain a second type of server having a load less than a preset value; a second control module, configured to control the fans deployed on the second type of servers to rotate at a speed lower than that of the fans deployed on the first type of servers; Or, a third control module is used to establish a connection between the fan deployed on the second type of server and the first type of server, and control the rotation of the fan deployed on the second type of server to reduce the temperature of the server components on the first type of server.

[0132] In some embodiments, the fan control device further comprises: A comparison module, configured to compare the current temperature value with a preset temperature value; wherein the preset temperature value is greater than the new current temperature threshold; a triggering module, configured to trigger a third acquiring module if the comparison result shows that the current temperature value is less than or equal to the preset temperature value; The control module is used to control the temperature of the server components through the liquid cooling mode if the comparison result shows that the current temperature value is greater than the preset temperature value, and then return to trigger the comparison module.

[0133] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, and they will not be repeated here.

[0134] Figure 5 This is a structural diagram of the server provided by the embodiment of the present invention. This embodiment is based on the hardware perspective, such as Figure 5 As shown, the server includes: Memory 20, for storing computer programs; The processor 21 is configured to implement the steps of the fan control method mentioned in the above embodiment when executing a computer program.

[0135] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented in at least one of the following hardware forms: a digital signal processor (DSP), a field-programmable gate array (FPGA), or a programmable logic array. The processor 21 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 21 may be integrated with a graphics processing unit (GPU), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 21 may also include an artificial intelligence (AI) processor, which is responsible for processing computing operations related to machine learning.

[0136] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201, wherein, after the computer program is loaded and executed by the processor 21, it can implement the relevant steps of the fan control method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include but is not limited to the data involved in the above-mentioned fan control method, etc.

[0137] In some embodiments, the server may further include a display screen 22 , an input / output interface 23 , a communication interface 24 , a power supply 25 , and a communication bus 26 .

[0138] Those skilled in the art will understand that Figure 5 The structure shown in the figure does not constitute a limitation to the server, and may include more or fewer components than shown in the figure.

[0139] The server provided by the embodiment of the present invention includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the following method: a fan control method, with the same effect as above.

[0140] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, which implements the steps of the above-mentioned fan control method when executed by a processor.

[0141] Finally, the present invention also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps described in the above method embodiment.

[0142] It is understood that if the methods in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0143] The computer-readable storage medium provided by the present invention includes the above-mentioned fan control method, and has the same effect as above.

[0144] The above is a detailed introduction to a fan control method, device, product, server and medium provided by the present invention. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, the present invention can also be improved and modified in several ways, and these improvements and modifications also fall within the scope of protection of the present invention.

[0145] It should also be noted that, in this specification, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A fan control method, characterized in that: include: Obtaining a current temperature threshold for indicating starting fan control, and obtaining historical temperature values ​​of server components under multiple ambient temperatures; Updating the current temperature threshold using the historical temperature value to obtain a new current temperature threshold; Obtaining a current temperature value of the server component; When it is detected that the current temperature value is greater than or equal to the new current temperature threshold, obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; The target speed value of the fan is outputted through the proportional-integral-differential control model, and the fan is controlled to rotate according to the target speed value.

2. The fan control method according to claim 1, wherein: Updating the current temperature threshold using the historical temperature value to obtain a new current temperature threshold includes: Get the peak temperature value in the historical temperature values; Performing weighted averaging processing on the temperature peak values ​​and obtaining a result after the weighted averaging processing; The new current temperature threshold is determined according to the current temperature threshold and the result after the weighted average processing.

3. The fan control method according to claim 1, wherein: Obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server includes: Obtaining control parameters of a proportional-integral-derivative control model, and obtaining a first temperature difference between the current temperature value and the new current temperature threshold; Establishing a proportional term, an integral term, and a differential term according to the control parameter and the first temperature difference; Determining a compensation coefficient corresponding to an item characterizing environmental compensation according to the ambient temperature value and the load condition of the server, and obtaining a second temperature difference between the current ambient temperature value and a reference ambient temperature value; establishing a term for characterizing environmental compensation according to the compensation coefficient and the second temperature difference; The proportional-integral-derivative control model is determined according to the proportional term, the integral term, the differential term, and the term for characterizing environmental compensation.

4. The fan control method according to claim 3, wherein: The compensation coefficients corresponding to the items used to characterize environmental compensation are determined based on the ambient temperature value and the server load condition, including: Get the load difference between the current load of the server and the maximum load supported by the server; Get the ratio between the load difference and the maximum load supported by the server; Obtaining a product of the rate of change of the ambient temperature and the ratio; The compensation coefficient corresponding to the item characterizing the environmental compensation is determined according to the preset reference compensation coefficient and the multiplication result.

5. The fan control method according to claim 3, wherein: The fan control mode includes at least a manual control mode, an automatic control mode through a proportional-integral-differential control model, or a control mode determined according to power consumption; wherein the manual control mode has the highest priority among all fan control modes.

6. The fan control method according to claim 5, wherein: After detecting that the current temperature value is greater than or equal to the new current temperature threshold, and before acquiring a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server, the method further includes: Get the current fan control mode; When it is detected that the current fan control mode is the automatic control mode using the proportional-integral-differential control model, entering the step of obtaining the proportional-integral-differential control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; When it is detected that the current fan control mode is the manual control mode, a rotation speed value input by a user is obtained; the rotation speed value input by the user is used as a target rotation speed value of the fan, and the fan is controlled to rotate according to the target rotation speed value.

7. The fan control method according to claim 5, wherein: The method further comprises: When it is detected that the current temperature value is less than the new current temperature threshold, determining the fan control mode to be a control mode determined according to power consumption; In the control mode determined according to the power consumption, the rotation speed value of the fan is controlled to be a preset rotation speed value; wherein the preset rotation speed value is smaller than the target rotation speed value.

8. The fan control method according to claim 7, wherein: Before detecting that the current temperature value is greater than or equal to the new current temperature threshold, the method further includes: Obtaining a first temperature difference between the current temperature value and the new current temperature threshold to determine a magnitude relationship between the current temperature value and the new current temperature threshold; Before obtaining a first temperature difference between the current temperature value and the new current temperature threshold, the method further includes: If a speed value input by the user is detected, determining that the current fan control mode is the manual control mode; using the speed value input by the user as the target speed value of the fan, and controlling the fan to rotate according to the target speed value; If the rotation speed value input by the user is not detected, the process proceeds to the step of obtaining the first temperature difference between the current temperature value and the new current temperature threshold.

9. The fan control method according to claim 8, wherein: After the fan speed is controlled to be a preset speed value, the method further includes: Acquire the ambient temperature value at each moment within a preset time period and the temperature value of the server component under the ambient temperature value to obtain the historical temperature value, and return to the step of updating the current temperature threshold using the historical temperature value.

10. The fan control method according to claim 8, wherein: Before obtaining the current temperature value of the server component, the method further includes: Obtain the historical temperature threshold corresponding to the speed manually set by the user and the frequency of manual user intervention; Determine the threshold weight based on the frequency of manual user intervention; A temperature threshold is determined according to the historical temperature threshold and the threshold weight to serve as the new current temperature threshold.

11. The fan control method according to claim 10, wherein: Before detecting the magnitude relationship between the current temperature value and the new current temperature threshold, the method further includes: determining a new current temperature threshold to be used; The determination of the new current temperature threshold to be used includes: A new current temperature threshold value obtained after the historical temperature value is updated to the current temperature threshold value within a preset time since the server starts processing the service is used as the new current temperature threshold value to be used; a temperature threshold value determined based on the historical temperature threshold value and the threshold weight is used as the new current temperature threshold value to be used after the preset time since the server starts processing the service; Or, get the current ambient temperature; Obtain the historical temperature conditions when the historical temperature value updates the current temperature threshold; When it is detected that the current ambient temperature is the same as the historical temperature, the new current temperature threshold obtained by updating the historical temperature value to the current temperature threshold is used as the new current temperature threshold; Alternatively, a new current temperature threshold value obtained by updating the current temperature threshold value with the historical temperature value is used as the initial temperature threshold value; After determining a temperature threshold according to the historical temperature threshold and the threshold weight, the initial temperature threshold is replaced by the temperature threshold determined according to the historical temperature threshold and the threshold weight to serve as a new current temperature threshold used.

12. The fan control method according to any one of claims 1 to 11, characterized in that: The obtaining of the current temperature value of the server component includes: Inputting historical temperature values ​​of the server components into a pre-established time series-based prediction model; The current temperature value of the server component is outputted through the time series-based prediction model.

13. The fan control method according to claim 12, wherein: After the target speed value of the fan is outputted through the proportional-integral-differential control model, and before the fan is controlled to rotate according to the target speed value, the method further includes: Determining a temperature jump frequency of the server component according to a temperature value of the server component predicted by the time series-based prediction model; Obtaining a jitter attenuation coefficient; wherein the jitter attenuation coefficient is determined by a temperature change trend; Determine a final target speed value according to the target speed value, the temperature jump frequency and the jitter attenuation coefficient; The controlling the fan to rotate according to the target speed value includes: Control the fan to rotate according to the final target speed value.

14. The fan control method according to claim 11, wherein: After outputting the target speed value of the fan through the proportional-integral-differential control model and controlling the fan to rotate according to the target speed value, the method further includes: Recording the ambient temperature value, the temperature value of the server component, the temperature threshold, the control parameter, and the compensation coefficient in a log; Copy the logs in the master control module to the backup control module; The method further comprises: When a failure of the main control module is detected, the temperature threshold data and the control parameters are restored according to the historical log data recorded in the backup control module; A simplified proportional-integral-differential control model is used to output a target speed value of the fan; wherein the simplified proportional-integral-differential control model includes a proportional term and an integral term; The fan is controlled to rotate according to the target speed value, and prompt information for indicating that the main control module is abnormal is output.

15. The fan control method according to claim 1, wherein: Also includes: Acquire a first type of server whose load is greater than or equal to a preset value, and acquire a second type of server whose load is less than the preset value; controlling the fan deployed on the second type of server to rotate at a speed lower than that of the fan deployed on the first type of server; Alternatively, a connection is established between the fan deployed on the second type of server and the first type of server, and the fan deployed on the second type of server is controlled to rotate to reduce the temperature of the server components on the first type of server.

16. The fan control method according to claim 1, wherein: After detecting that the current temperature value is greater than or equal to the new current temperature threshold, and before obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server, the method further includes: Comparing the relationship between the current temperature value and the preset temperature value; wherein the preset temperature value is greater than the new current temperature threshold; If the comparison result is that the current temperature value is less than or equal to the preset temperature value, then entering the step of obtaining a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; If the comparison result is that the current temperature value is greater than the preset temperature value, after regulating the temperature of the server component through the liquid cooling mode, the process returns to the step of comparing the relationship between the current temperature value and the preset temperature value.

17. A fan control device, characterized in that: include: A first acquisition module is used to acquire a current temperature threshold for indicating starting fan control, and to acquire historical temperature values ​​of server components under multiple ambient temperatures; An updating module, configured to update a current temperature threshold using the historical temperature value to obtain a new current temperature threshold; A second acquisition module is used to obtain the current temperature value of the server component; a third acquisition module, configured to, when detecting that the current temperature value is greater than or equal to the new current temperature threshold, acquire a proportional-integral-derivative control model determined by the new current temperature threshold, the current temperature value, the ambient temperature value, and the load condition of the server; The output and control module is used to output the target speed value of the fan through the proportional-integral-differential control model and control the fan to rotate according to the target speed value.

18. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the fan control method according to any one of claims 1 to 16 are implemented.

19. A server, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the fan control method according to any one of claims 1 to 16 when executing the computer program.

20. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the fan control method according to any one of claims 1 to 16 are implemented.

Images