Temperature value verification method and computing device

Abstract

The embodiment of the invention provides a temperature value verification method and computing equipment. The method comprises the following steps: acquiring a first temperature value of a target component and a second temperature value of a to-be-measured component; determining a third temperature value and a reference temperature difference of the to-be-measured component according to the first temperature value and a pre-constructed temperature estimation model; the temperature estimation model is used for representing an association relationship between a first temperature value of the target component, a third temperature value of the to-be-measured component and a reference temperature difference under different environment temperatures and business pressures; the environment temperature indicates the temperature value of the external environment when the to-be-tested component operates; the service pressure indicates the size of the working load when the to-be-tested part bears the service to run; and determining whether the second temperature value is an abnormal temperature value based on the temperature difference between the second temperature value and the third temperature value and the reference temperature difference. According to the scheme, whether the obtained temperature value of the component is the abnormal temperature value or not can be checked in an auxiliary mode, and therefore the problem that the component falls off due to over-temperature caused by the abnormal temperature value is avoided.

Description

Technical Field

[0001] The present application relates to the field of server technology, and in particular to a temperature value verification method and computing equipment. Background Art

[0002] At present, fan cooling is a common cooling solution for the internal components of servers. For example, in a server, the baseboard management controller (BMC) manages the cooling of compatible components by obtaining the corresponding temperature from the components through a protocol, and then controls the speed of the corresponding fan to achieve heat dissipation and cooling of the corresponding components.

[0003] As components become more complex, especially when they have their own management software, once the component management software fails, it may return an abnormal temperature value to the server BMC. This may cause the server BMC to incorrectly judge the temperature status of the component, thereby affecting the effect of fan speed regulation. If the fan speed is not properly regulated, the component may stop working normally due to overheating. For example, when the component exceeds its safe operating temperature range, it will automatically reduce performance or stop running (also known as over-temperature card failure), affecting business use. Summary of the invention

[0004] A temperature value verification method and a computing device are provided in an embodiment of the present application, which can assist in verifying whether the temperature value of a component obtained is an abnormal temperature value, thereby avoiding the problem of overheating and disconnection of the component due to abnormal temperature values.

[0005] To this end, the following technical solutions are provided in the embodiments of the present application:

[0006] In a first aspect, an embodiment of the present application provides a temperature value verification method, comprising: obtaining a first temperature value of a target component and a second temperature value of a component to be measured; wherein the target component and the component to be measured have a correlation, and the correlation indicates that the size of the first temperature value can affect the size of the second temperature value; determining a third temperature value and a reference temperature difference of the component to be measured according to the first temperature value and a pre-constructed temperature estimation model; the temperature estimation model is used to represent the association between the first temperature value of the target component, the third temperature value of the component to be measured and the reference temperature difference under different ambient temperatures and business pressures; the ambient temperature indicates the temperature value of the external environment when the component to be measured is running; the business pressure indicates the size of the workload of the component to be measured when carrying out business operation; based on the temperature difference between the second temperature value and the third temperature value and the reference temperature difference, confirming whether the second temperature value is an abnormal temperature value.

[0007] In this embodiment, the target component related to the component to be measured is first selected as a reference point, which ensures that the reference temperature (first temperature value) obtained is closely related to the actual situation of the component to be measured. Therefore, the target component can be used as the temperature reference point of the component to be measured to estimate the temperature of the component to be measured, which is conducive to reducing the difference between the estimated temperature value (third temperature value) and the actual temperature value (second temperature value), thereby enhancing accuracy and reliability. Afterwards, the establishment of the temperature estimation model is to construct a temperature estimation model by summarizing and fitting the actual temperature of the component to be measured and the reference point temperature under different parameter conditions. This step takes into account the influence of different working parameter conditions on the temperature, so that the model can more accurately predict the expected temperature of the component to be measured. In the verification process, the concept of reference temperature difference is introduced to determine whether the difference between the third temperature value and the second temperature value returned by the component to be measured is within an acceptable range, so that abnormal temperature values ​​can be effectively identified and the possibility of misjudgment can be reduced. In this way, the fan corresponding to the component can dissipate heat for the component according to the normal temperature value, thereby avoiding the problem of overheating and card dropping due to abnormal temperature values.

[0008] Thus, the embodiment of the present application improves the system's ability to detect and respond to component temperature anomalies by comprehensively utilizing the temperature sensor data of the target component as the temperature reference point, summarizing the temperature estimation model obtained by fitting, and the reference temperature difference error control strategy. This comprehensive performance advantage not only improves the stability and reliability of the system, but also reduces service interruption or degradation caused by temperature anomalies.

[0009] As an achievable implementation, the method further includes: selecting a component that has the smallest distance from the component to be measured and is provided with a temperature sensor as the target component.

[0010] In this embodiment, the target component that is closest to the component to be measured is selected as the temperature reference point, which ensures that the acquired temperature data is closest to the actual situation of the component to be measured, and the correlation between the two is closer. This accuracy and reliability ensures that the temperature reference point can accurately reflect the actual operating temperature of the component to be measured, avoiding temperature measurement deviations caused by excessive distance. In addition, using the existing target component as a temperature reference point avoids additional sensor installation and debugging costs. This not only simplifies the implementation process of the technical solution, but also reduces the cost of system maintenance and operation, and improves the overall cost-effectiveness ratio. Moreover, by selecting the target component as the temperature reference point, the technical solution demonstrates adaptability and scalability under different server configurations and layouts, ensuring that each component to be measured can have its optimal temperature reference point.

[0011] As an achievable implementation, the target component may also be a temperature sensor pre-configured on the side of the component to be measured, and the pre-configured temperature sensor is used as the temperature reference point of the component to be measured. It should be noted that the temperature reference point is as close to the component to be measured as possible, and the temperature reference point needs to be reserved during hardware design.

[0012] In this embodiment, a pre-configured temperature sensor is used as an alternative solution to ensure that even when the target component is far away from the component to be tested, the necessary temperature data of the temperature reference point can still be obtained. This method ensures that necessary temperature monitoring and verification management can be performed in various situations, thereby maintaining the safety and stability of component operation.

[0013] As an achievable implementation method, the temperature estimation model includes a lookup table of actual temperature values ​​of the component to be measured and a reference temperature difference, and the method also includes: obtaining the actual temperature value of the component to be measured when working normally under different business pressures within a preset working environment temperature range and the reference temperature value of the target component; summarizing and combining the actual temperature value of the component to be measured and the reference temperature value of the target component under different ambient temperatures and business pressures, and setting the allowable difference between the actual temperature value and the reference temperature value under different combinations as the reference temperature difference, thereby forming a lookup table of actual temperature values ​​of the component to be measured and a reference temperature difference.

[0014] In this embodiment, the ambient temperature and business pressure are one of the important factors that affect the temperature of the internal components of the server. By comprehensively considering the two parameter conditions of ambient temperature and business pressure, the expected temperature of the component to be tested can be predicted more accurately, which improves the accuracy of temperature prediction, and can accurately identify the situation where the self-test temperature value of the component is an abnormal temperature value, which helps to avoid incorrect speed regulation or overheating caused by external temperature changes or increased business pressure, and ensures the safe operation and stability of the internal components of the server under different working conditions. Considering the impact of changes in ambient temperature and business pressure on the temperature of the component to be tested, by making detailed arrangements and combinations of the ambient temperature and business pressure that affect the temperature of the component to be tested, and obtaining actual measurement data at the same time, and by summarizing and integrating the data under different conditions, it is possible to fully understand how these factors interact and affect temperature changes, so that an accurate temperature estimation model can be established, and such a model can accurately estimate the expected temperature value of the component to be tested under different workloads and environmental conditions.

[0015] As an achievable implementation method, the third temperature value and reference temperature difference of the component to be tested are determined based on the first temperature value and a pre-built temperature estimation model, including: obtaining the current business pressure parameters and working environment temperature of the component to be tested; searching for the actual temperature value and reference temperature difference of the component to be tested corresponding to the combination of the current business pressure parameters, working environment temperature and the first temperature value in a lookup table of actual temperature values ​​and reference temperature differences of the component to be tested; wherein the actual temperature value searched for is the third temperature value.

[0016] In this embodiment, by acquiring the current ambient temperature, business pressure and real-time temperature information of the temperature reference point and inputting them into the temperature estimation model, the expected temperature value (third temperature value) of the component to be tested can be calculated in real time.

[0017] As an achievable implementation method, the pre-construction of the temperature estimation model includes: obtaining the actual temperature value of the component to be measured when it works normally under different business pressures within a preset working environment temperature range and the reference temperature value of the target component; summarizing and fitting the actual temperature value of the component to be measured and the reference temperature value of the target component under different ambient temperatures and business pressures, and setting the allowable difference between the actual temperature value and the reference temperature value under different combinations as the reference temperature difference, thereby obtaining the functional relationship between the actual temperature value of the component to be measured and the reference temperature difference.

[0018] As an achievable implementation method, the third temperature value and reference temperature difference of the component to be tested are determined based on the first temperature value and a pre-built temperature estimation model, including: obtaining the current business pressure parameters and the working environment temperature of the component to be tested; determining the actual temperature value and the reference temperature difference of the component to be tested based on the functional relationship between the actual temperature value and the reference temperature difference of the component to be tested and the current business pressure parameters and the working environment temperature; wherein the determined actual temperature value is the third temperature value.

[0019] As an achievable implementation manner, the method further includes: generating alarm information in response to a temperature difference between the second temperature value and the third temperature value being greater than the reference temperature difference.

[0020] In this embodiment, by comparing the difference (temperature difference) between the expected temperature value (third temperature value) and the self-test temperature value (second temperature value) in real time, the server BMC can detect temperature anomalies immediately. Once the difference exceeds the preset error range, an alarm message can be immediately generated to notify the operator or automatically trigger corresponding emergency measures. This real-time monitoring and early warning capability greatly improves the system's sensitivity and response speed to the temperature status of components, and helps to quickly prevent and resolve risks and potential losses that may be caused by temperature anomalies. For example, when the temperature of a component exceeds the safe range, the system can immediately take measures, such as reducing the load, increasing the fan speed, etc., to avoid damage or shutdown caused by overheating of the component. This early warning mechanism not only helps to protect the safety and reliability of the equipment, but also helps to protect business continuity and user experience.

[0021] As an achievable implementation, the method also includes: when confirming that the second temperature value is an abnormal temperature value, controlling the fan to perform abnormal speed regulation; wherein the abnormal speed regulation indicates that the fan is running at a constant speed value, and when the fan is running at a constant speed value, the component to be tested can operate normally.

[0022] In this embodiment, when it is confirmed that the actual temperature value of the component to be tested is abnormal, abnormal speed regulation will be triggered, that is, the speed of the fan will be adjusted to cool the component more effectively. For example, the constant speed value of the fan can be the maximum speed, which can help avoid the situation where the component stops working normally or the performance decreases due to overheating, thereby ensuring the stable operation of the component.

[0023] As a feasible implementation method, the components to be tested include but are not limited to smart network cards, network card optical modules, GPU modules, NPU modules and other components that need to obtain temperature through software protocols.

[0024] In this implementation, not all compatible components need to be checked for abnormal temperature values. During the design phase, critical components that are sensitive to temperature rise and highly dependent on software to obtain temperature are evaluated and this technical solution is implemented. For the above-mentioned high-performance components such as smart network cards, network card optical modules, NPU modules, GPU modules, etc., they play a key role in high-performance computing, data centers or network equipment. Accurately controlling their operating temperature is an important factor in ensuring performance and reliability. By obtaining the ambient temperature and business pressure in real time, and monitoring and adjusting based on the expected temperature value, the above-mentioned modules can be effectively avoided from overheating, thereby improving the stability and life of each component in the above-mentioned module.

[0025] As a feasible implementation manner, the component to be tested is a smart network card, and the target component is a baseboard management controller.

[0026] As a feasible implementation manner, the component to be tested is a smart network card, and the target component is a pre-configured reference temperature sensor.

[0027] As an achievable implementation, the second temperature value of the component to be tested can be obtained by the following method: obtaining the self-test temperature value of the network card optical module through the mctp protocol; obtaining the self-test temperature value of each component to be tested of the NPU module and the GPU module through the redfish protocol; obtaining the self-test temperature value of each component from the smart network card through the ipmi protocol. It can be understood that the above self-test temperature value is the second temperature value.

[0028] In a second aspect, an embodiment of the present application further provides a temperature value verification device, comprising:

[0029] An acquisition module, used for acquiring a first temperature value of a target component and a second temperature value of a component to be measured; wherein the target component and the component to be measured have a correlation, and the correlation indicates that the magnitude of the first temperature value can affect the magnitude of the second temperature value;

[0030] A temperature determination module, used to determine the third temperature value and the reference temperature difference of the component to be tested according to the first temperature value and a pre-built temperature estimation model; wherein the temperature estimation model is used to indicate the correlation between the first temperature value of the target component, the third temperature value of the component to be tested and the reference temperature difference under different ambient temperatures and business pressures; the ambient temperature indicates the temperature value of the external environment when the component to be tested is running; the business pressure indicates the size of the workload of the component to be tested when carrying business operations;

[0031] The verification module is used to confirm whether the second temperature value is an abnormal temperature value based on whether the temperature difference between the second temperature value and the third temperature value is greater than the reference temperature difference.

[0032] As an achievable implementation, the device further includes a preprocessing module, and the preprocessing module is used to select a component that has the smallest distance from the component to be measured and is provided with a temperature sensor as a target component.

[0033] As an achievable implementation manner, the preprocessing module is further used to select a temperature sensor pre-configured on the side of the component to be measured as the target component, and use the pre-configured temperature sensor as a temperature reference point of the component to be measured.

[0034] As an achievable implementation method, the preprocessing module is also used to pre-build the temperature estimation module, including: determining a component that has the smallest distance from the component to be measured and is equipped with a temperature sensor as a target component; obtaining an actual temperature value of the component to be measured when working normally under different business pressures within a preset working environment temperature range and a reference temperature value of the target component; summarizing and combining the actual temperature values ​​of the component to be measured and the reference temperature values ​​of the target component under different ambient temperatures and business pressures, and setting the allowable difference between the actual temperature value and the reference temperature value under different combinations as the reference temperature difference, thereby forming a lookup table of the actual temperature value and the reference temperature difference of the component to be measured.

[0035] As an achievable implementation method, the acquisition module is also used to obtain the current business pressure parameters and working environment temperature of the component to be tested; in the actual temperature value and reference temperature difference lookup table of the component to be tested, the actual temperature value and reference temperature difference of the component to be tested corresponding to the combination of the current business pressure parameters, working environment temperature and the first temperature value are searched; wherein the actual temperature value searched is the third temperature value.

[0036] As an achievable implementation method, the preprocessing module is also used to pre-build the temperature estimation module, including: determining a component that has the smallest distance from the component to be measured and is equipped with a temperature sensor as a target component; obtaining an actual temperature value of the component to be measured when working normally under different business pressures within a preset working environment temperature range and a reference temperature value of the target component; summarizing and fitting the actual temperature value of the component to be measured and the reference temperature value of the target component under different ambient temperatures and business pressures, and setting the allowable difference between the actual temperature value and the reference temperature value under different combinations as the reference temperature difference, thereby obtaining a functional relationship between the actual temperature value of the component to be measured and the reference temperature difference.

[0037] As an achievable implementation method, the acquisition module is also used to obtain the current business pressure parameters and working environment temperature of the component to be tested; determine the actual temperature value and the reference temperature difference of the component to be tested based on the functional relationship between the actual temperature value and the reference temperature difference of the component to be tested and the current business pressure parameters and the working environment temperature; wherein the determined actual temperature value is the third temperature value.

[0038] As an achievable implementation, the temperature value verification device further includes: a processing module, configured to generate an alarm message in response to the temperature difference between the second temperature value and the third temperature value being greater than the reference temperature difference.

[0039] As an achievable implementation, the processing module is also used to trigger abnormal fan speed regulation when it is confirmed that the second temperature value is an abnormal temperature value; wherein, the abnormal speed regulation indicates that the fan is running at a constant speed value, and when the fan is running at a constant speed value, the component to be tested can operate normally.

[0040] In a third aspect, an embodiment of the present application further provides a computing device, the computing device comprising a baseboard management controller, the baseboard management controller being used to execute the method involved in the first aspect and possible implementations thereof.

[0041] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions in the computer-readable storage medium are executed by a computing device, the computing device executes the method involved in the first aspect and its possible implementation methods.

[0042] In a fifth aspect, an embodiment of the present application further provides a computer program product, which includes computer instructions. When the computer instructions are executed by a computing device, the computing device executes the method described in the first aspect and its possible implementation methods.

[0043] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0044] In summary, the embodiments of the present application effectively solve the fan speed adjustment problem caused by abnormal component temperature in the server by introducing a temperature value verification method and a computing device. Its advantages lie in the accurate temperature estimation model, reliable error range control, and fast abnormal speed adjustment response, which jointly guarantee the stability and performance of computing devices such as server systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of the structure of a computing device provided in an embodiment of the present application is shown;

[0046] Figure 2 A schematic diagram of a temperature value verification method provided in an embodiment of the present application is shown;

[0047] Figure 3 A temperature value verification device provided by an embodiment of the application is shown;

[0048] Figure 4 A schematic structural diagram of an example of a computing device is shown. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0050] In the description of the embodiments of the present application, words such as "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a concrete way.

[0051] In the description of the embodiments of the present application, the term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, B exists alone, and A and B exist at the same time. In addition, unless otherwise specified, the term "multiple" means two or more. For example, multiple systems refers to two or more systems, and multiple terminals refers to two or more terminals.

[0052] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprises", "has" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0053] In the description of the embodiments of the present application, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it can be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0054] In the description of the embodiments of the present application, the terms "first\second\third, etc." or module A, module B, module C, etc. are only used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that the specific order or sequence can be interchanged where permitted so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0055] In the description of the embodiments of the present application, the numbers representing the steps, such as S31, S32, etc., do not necessarily mean that the steps will be executed in this manner. If permitted, the order of the previous and next steps can be interchanged, or they can be executed simultaneously.

[0056] Relevant terms involved in the embodiments of this application:

[0057] BMC, baseboard management controller (BMC), is a control chip located inside the server, responsible for the server's component status management, such as component temperature, humidity, power supply voltage and fan speed detection; operating system management, health status management, function consumption management and other core functions.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0059] To facilitate understanding of the technical solution of the present application, a computing device provided in an embodiment of the present application is introduced below.

[0060] Figure 1 A schematic diagram of the structure of a computing device provided in an embodiment of the present application is shown. Figure 1 As shown, an embodiment of the present application provides a computing device, which may be a server or a terminal, and includes: a BMC and a fan. The fan controls the speed of the fan according to the fan control signal output by the BMC. Exemplarily, the fan control signal may be a PWM (pulse width modulation) signal. It is understandable that PWM is a technology that uses pulses to output analog signals, and is a method of digitally encoding the level of an analog signal, by modulating the duty cycle of a square wave, thereby encoding the level of an analog signal. In a cooling fan, PWM can be used to control the speed of the fan, for example, by inputting a PWM signal to the control end of the fan, and adjusting the speed of the fan by adjusting the duty cycle of the PWM wave, thereby adjusting the temperature of the central processing unit CPU or other components inside the computing device.

[0061] It should be noted that Figure 1 It is only an example of a computing device and does not constitute a specific limitation. In practical applications, it may include Figure 1More or fewer other components, such as other components may also include a single-chip microcomputer (an integrated circuit chip, equivalent to a microcomputer), a digital signal processor (DSP), an application specific integrated circuit (ASIC), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, general-purpose processors, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. In addition, in another possible implementation, other components may also include other compatible components such as any one of a smart network card, a network card optical module, a NPU (neural processing unit), and a GPU (graphics processing unit).

[0062] like Figure 1 As shown, in a possible implementation, the BMC determines the fan control signal based on the temperature information of other components in the computing device. Exemplarily, the BMC obtains the temperature information of the components in the computing device through temperature sensors set at the positions of other components in the computing device. The BMC will calculate the required fan speed based on the temperature data obtained by the sensor, and send a corresponding control signal (such as a PWM signal) to the fan control circuit to adjust the fan speed. The BMC can also be electrically connected to other components in the computing device to perform other tasks. It should be noted that heat dissipation of components in the computing device is a common scenario for the BMC to control the speed of the fan body. Other scenarios may be testing fan performance, performing pressure resistance tests on the fan, etc.

[0063] In a possible implementation, other components in the computing device may also include a complex programmable logic device (CPLD), which is electrically connected to the BMC. In an embodiment of the present application, the CPLD can receive a control signal sent by the BMC, and output a PWM signal with different duty cycles according to the control signal. In addition to the BMC directly generating a fan control signal, the BMC generating a fan control signal through the CPLD is another implementation of the BMC generating a fan control signal. In other implementations, the BMC generating a fan control signal can also be implemented by hardware such as a field programmable gate array (FPGA) and a generic array logic (GAL).

[0064] In other words, the BMC can be electrically connected to the fan control chip and directly send a fan control signal to the fan control chip. The BMC can also be electrically connected to hardware such as CPLD, FPGA, GAL, etc. The CPLD, FPGA, GAL, etc. hardware are electrically connected to the fan control chip. The BMC controls the CPLD, FPGA, GAL, etc. hardware to generate a fan control signal and control the corresponding fan to run at the corresponding speed.

[0065] The following uses a server as an example to explain in detail how the BMC implements heat dissipation management for other components.

[0066] like Figure 1 As shown, the server manages the fan through the BMC. The BMC formulates a speed control strategy for the fan in the server and monitors the temperature information of the corresponding positions of each component in the server. Under the guidance of the control strategy, the fan is controlled to produce different speeds to ensure that the temperature in the server environment is equal to or less than the set temperature threshold. Once the overtemperature condition is triggered, the BMC will adjust the duty cycle of the fan's PWM control signal through the CPLD (increase or decrease), thereby increasing the fan speed. Specifically, the following steps are included:

[0067] Step 1: The BMC sets an initial speed for the fans in the server. Initially, the speeds of the fans are the same.

[0068] Step 2: The BMC determines whether the speed of the fan in the server needs to be adjusted based on the self-test temperature value sent back by the temperature sensor at the corresponding position (eg, built-in) of each component in the server and the working condition (eg, workload) of each component in the server.

[0069] It is understandable that BMC generally obtains the corresponding temperature value from the component through a certain protocol and performs intelligent speed regulation. For example, BMC obtains the self-test temperature value of the network card optical module through the mctp protocol; obtains the self-test temperature value of each component of the NPU, GPU module, and atlas module through the redfish protocol; and obtains the self-test temperature value of each component from the smart network card through the ipmi protocol.

[0070] Step 3: When the temperature of a certain location or component in the server exceeds a preset temperature threshold (e.g., 50°C), or a certain component in the server is undergoing a stress test, the BMC controls the CPLD to adjust the duty cycle of the PWM output signal, and the CPLD sends the PWM signal to the fan corresponding to the location, thereby increasing the fan speed, thereby reducing the temperature of the location or component inside the server.

[0071] The embodiment of the present application provides a temperature value verification method, which uses an external temperature sensor as a reference point to fit a component temperature reference value to verify the self-test temperature value of the component obtained by the component management software. It can assist in verifying whether the temperature value of the obtained component is an abnormal temperature value. When the verification fails, abnormal speed regulation is triggered, thereby avoiding the problem of overheating and card disconnection due to abnormal temperature values ​​of the component, and increasing the reliability of intelligent speed regulation of the server.

[0072] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. The several specific embodiments listed can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0073] Next, refer to Figure 1 The server shown introduces a temperature value verification method provided in an embodiment of the present application.

[0074] Figure 2 A flow chart of a temperature value verification method provided by an embodiment of the present application is shown. The temperature value verification method is suitable for execution in a computing device (such as the aforementioned server), and is used to verify the temperature value of a compatibility component in the server that performs heat dissipation management through a BMC, wherein the compatibility component includes a hardware component installed in the same server as the BMC, and the hardware component includes a component to be tested and a target component. Figure 2 As shown, the temperature value verification method includes:

[0075] S101, obtaining a first temperature value of a target component and a second temperature value of a component to be measured, wherein the target component and the component to be measured are correlated, and the correlation indicates that the magnitude of the first temperature value can affect the magnitude of the second temperature value.

[0076] It should be noted that not all components need to be checked for abnormal temperature values, so it is necessary to evaluate key components that are sensitive to temperature rise and strongly rely on BMC to obtain temperature during the design phase. For example, the components to be tested may include any compatible components in smart network cards, network card optical modules, NPU modules, and GPU modules. Exemplarily, the BMC can obtain the second temperature value (self-test temperature value) returned by the component to be tested under current conditions through a software protocol. For example, the BMC obtains the self-test temperature values ​​of each component of the NPU module and the GPU module through the redfish protocol; the BMC obtains the self-test temperature values ​​of each component from the smart network card through the ipmi protocol; the BMC obtains the self-test temperature value of the network card optical module through the mctp protocol.

[0077] It should be understood that the above-mentioned components to be tested are only exemplary illustrations. In some embodiments, technical personnel in this field can also select any other components to implement the present technical solution according to design requirements. This implementation is not strictly limited here. For the temperature anomaly value verification of any other components, the technical solution provided in the embodiment of the present application is also applicable.

[0078] After determining the component to be tested that needs temperature estimation, a temperature sensor that is relevant to the component to be tested is searched in the internal structure of the server. Exemplarily, a component with the smallest distance from the component to be tested and equipped with a temperature sensor is selected as the target component, ensuring that the acquired temperature data is closest to the actual situation of the component to be tested. In this way, the temperature sensor on the target component is used as a temperature reference point. Since the distance to the component to be tested is the shortest, the temperature reference point can more accurately reflect the actual operating temperature of the component to be tested, avoiding temperature measurement deviations caused by excessive distance. Moreover, by using the target component as a temperature reference point, the adaptability and scalability under different server configurations and layouts are demonstrated, ensuring that each component to be tested has its optimal temperature reference point without the need to pre-configure the temperature sensor.

[0079] In another embodiment, the target component may also be a temperature sensor pre-configured on the side of the component to be measured, and the pre-configured temperature sensor is used as the temperature reference point of the component to be measured. It should be noted that the temperature reference point is as close to the component to be measured as possible, and the temperature reference point needs to be reserved during hardware design. It is understandable that in the case where the compatible component equipped with the temperature sensor is far away from the component to be measured, in order to make the temperature reference point of the component to be measured more accurate, it is necessary to consider reserving a position for configuring the temperature sensor within a preset range during hardware design. In this way, the temperature sensor pre-configured on the side of the component to be measured can be as close to the component to be measured as a temperature reference point, and the necessary temperature data of the temperature reference point can still be obtained.

[0080] In other words, the temperature sensor that is correlated with the component to be measured can be the temperature sensor that is closest to the component to be measured. The pre-configured temperature sensor is used as the temperature reference point of the component to be measured. The pre-configured temperature sensor is closest to the component to be measured, which can improve the accuracy of the temperature value verification of the component to be measured. It can be understood that the pre-configured temperature sensor can be arranged according to the specific system design and deployment requirements, and is not limited to the specific electronic component position or layout. In the case where the server or device layout is complex, the electronic component position is not fixed or difficult to access, the pre-configured temperature sensor enables the server BMC to adjust the position of the temperature reference point according to different deployment situations and requirements to ensure the adaptability and accuracy of temperature verification. Alternatively, the target component where the temperature sensor that is correlated with the component to be measured is located is on the equivalent circuit of the same connection line as the component to be measured. For example, the electronic components in the power supply unit (PSU) and the voltage regulator module (VRM) on the motherboard, the memory module (RAM) and the memory controller, the hard disk and the hard disk drive, etc. The above modules are on the same circuit connection line and have correlation. The above target components and the component to be measured will show correlation in the working state. If the temperature of the memory module rises, the temperature of the memory controller in the same connection line equivalent circuit will also rise. In this scenario, correlation refers to whether the state variables of the target component and the state variables of the component to be tested can be converted and influence each other. For example, the loss of a certain target component is positively correlated with the real-time input current of the server. The higher the real-time input current, the higher the loss of the target component, and the higher the temperature that can be detected by the temperature sensor set on the target component. The loss of the component to be tested is also positively correlated with the real-time input current of the server. The higher the real-time input current, the higher the loss of the component to be tested. Therefore, it can be obtained that the loss of the target component and the loss of the component to be tested increase or decrease almost at the same time. According to conventional settings, the higher the loss of the target component, the higher the loss of the component to be tested. At this time, the temperature that can be detected by the target component and the component to be tested is also higher, thereby determining that the state variable of the target component is associated with the state variable of the component to be tested.

[0081] The following uses the power module and the voltage regulation module as examples to illustrate the correlation between the state variables of the two.

[0082] Since the electronic components such as inductors, capacitors, transformers in the power module are directly related to the voltage regulating module on the mainboard, when the voltage regulating module is used as the component to be tested for temperature value verification, the electronic components with temperature sensors on the power module that are related to the voltage regulating module can be used as temperature reference points. This is because the higher the input current of the power module, the higher the internal loss and temperature increase will affect the stability and efficiency of the voltage regulating module, causing the temperature of the voltage regulating module to gradually increase. In addition, the voltage regulating module is responsible for stably outputting the power supply voltage to the processor and other components, so the state and temperature of the power module and the voltage regulating module VRM on the circuit are interrelated. In other words, the correlation is determined based on the relevant state information of the component to be tested and the relevant state information of the temperature sensor on the target component, so the relevant state information between the component to be tested and the temperature sensor can be converted and influenced by each other, so the temperature sensor is used as the temperature reference point of the component to be tested, which can reduce the temperature difference between the expected temperature value of the component to be tested and the actual temperature without the need to set up a new temperature sensor, thereby improving the accuracy of the temperature verification of the component to be tested.

[0083] S102: Determine a third temperature value and a reference temperature difference of the component to be tested according to the first temperature value and a pre-built temperature estimation model, wherein the reference temperature difference indicates the maximum difference between the second temperature value and the third temperature value.

[0084] The temperature estimation model is established based on the temperature values ​​of the target component and the component to be tested when they are working normally under different business pressures within the preset working environment temperature range; the temperature estimation model is used to express the correlation between the first temperature value of the target component, the third temperature value of the component to be tested and the reference temperature difference under different environmental temperatures and business pressures. The environmental temperature indicates the temperature value of the external environment when the component to be tested is running; the business pressure indicates the size of the workload of the component to be tested when carrying business operations.

[0085] It should be noted that different parameter conditions have a greater impact on the temperature of the component to be tested, for example:

[0086] Ambient temperature: Ambient temperature directly affects the heat dissipation effect of the component under test. A higher ambient temperature will make it more difficult for the component to dissipate heat, thereby causing the component temperature to rise. It is understandable that the ambient temperature can be the temperature inside the server chassis or the temperature of the environment outside the server chassis.

[0087] Business pressure: The power consumption and heat generation of components will vary under different business pressure (workload) conditions. Under high business pressure conditions, the power consumption of components increases and the heat generated also increases.

[0088] Power supply voltage: The stability and voltage level of the input voltage will affect the power consumption and heat generation of the component under test. Excessive or unstable voltage may cause increased heating of the component.

[0089] Component layout: The layout of components in the device affects the distribution and heat dissipation efficiency of heat. Tight layout may cause heat accumulation and increase the temperature.

[0090] Component aging: As components age, their performance and heat dissipation capabilities may decline, causing the temperature to rise.

[0091] External interference: External factors such as heat from surrounding equipment may also affect the temperature of the component under test.

[0092] Power density: The power density of a component, i.e. the power consumption per unit area, directly affects the heat generation. Components with high power density usually have higher temperatures.

[0093] By considering these factors and summarizing and fitting the experimental data, a more accurate temperature estimation model can be established. These factors are usually reflected in the temperature estimation model in the form of weights or correction factors to comprehensively influence the comprehensive effect of various conditions on temperature.

[0094] Exemplarily, in this embodiment, the parameter conditions selected for establishing the temperature estimation model include ambient temperature and business pressure. The ambient temperature indicates the temperature value of the external environment when the component under test is running. The business pressure indicates the size and intensity of the workload of the component under test when carrying the business. Ambient temperature and business pressure are one of the important factors affecting the temperature of the internal components of the server. By comprehensively considering the two parameter conditions of ambient temperature and business pressure, a temperature estimation model that more accurately predicts the expected temperature of the component under test can be established.

[0095] The following is a detailed description of the method for establishing the temperature estimation model. The temperature estimation model can be established in the following two ways:

[0096] The first method is to establish a temperature estimation model by establishing a lookup table. Specifically, establishing the temperature estimation model includes the following steps:

[0097] S1021. Obtain an actual temperature value of the component to be tested within a preset working environment temperature range when it works normally under different business pressures and a reference temperature value of the target component.

[0098] Optionally, developers can obtain different environmental temperatures and business pressures during the server business stress test during the development and design phase.

[0099] For example, a temperature sensor is installed inside the server chassis to monitor the real-time internal temperature. In another possible implementation, if it is necessary to understand the impact of the room temperature on the component temperature, an ambient temperature sensor can be installed in the room.

[0100] Exemplarily, stress testing tools (such as Apache JMeter, LoadRunner, Gatling, Locust, and Artillery, etc.) are used to simulate different business stress situations, and the temperature changes of components under different stresses are recorded.

[0101] In the laboratory, based on factors that affect temperature changes, such as ambient temperature and business pressure, the reference temperature values ​​of the temperature reference points and the actual temperature values ​​of the components to be tested under different combinations of ambient temperature and business pressure are actually tested.

[0102] For example, when the preset working environment temperature range of the component to be tested is 0°C to 50°C, the test can be started from 0°C, and 1°C is added as a node as a temperature collection point, that is, the environmental temperature collection points of the component to be tested are 0°C, 1°C, 2°C, 3°C, 4°C, 5°C... 47°C, 48°C, 49°C, 50°C. In different working environments, the business pressure of the component to be tested is gradually increased, for example, the workload of the component to be tested is adjusted from 0% to 100%, and the increase can be 1%. In this embodiment, there is no strict restriction on the increase in the working environment temperature and workload.

[0103] In one example, the BMC interface can be used to obtain the actual temperature value of the component under test at different ambient temperatures and the reference temperature value of the temperature reference point. Optionally, developers can also arrange multiple temperature sensors on the component under test to obtain the actual temperature value of the component under test, to avoid a single temperature sensor failing to obtain the actual temperature value of the component under test due to a failure.

[0104] S1022. The actual temperature values ​​of the components to be tested and the reference temperature values ​​of the target components under different ambient temperatures and business pressures are summarized and combined, and the allowable differences between the actual temperature values ​​and the reference temperature values ​​under different combinations are set as reference temperature differences, thereby forming a lookup table of actual temperature values ​​and reference temperature differences of the components to be tested.

[0105] It should be understood that in this step, the establishment of the temperature estimation model is conducive to the rapid verification of the temperature value. Moreover, the temperature estimation model can not only output the predicted value of the actual temperature, but also quantify the error range (reference temperature difference) between the expected temperature value and the measured temperature value. This precise quantification helps to evaluate the accuracy and reliability of whether the self-test temperature value returned by the component is an abnormal temperature value, thereby providing accurate temperature verification for the normal operation of the component.

[0106] Exemplarily, in this implementation, the temperature estimation model is a lookup table, as shown in Table 1:

[0107] Table 1 Example of forming a lookup table for different ambient temperatures, service pressures, actual temperature values ​​of the components to be tested, and reference temperature values ​​of the temperature reference points

[0108] Ambient temperature Business Pressure Reference point temperature value Component temperature value Allowable error range (reference temperature difference) / ± 0℃ 0％ 2℃ 3℃ 1℃ 0℃ 1％ 2℃ 3℃ 1℃ 0℃ 2％ 2℃ 3℃ 1℃ 0℃ 3％ 2℃ 3℃ 1℃ 0℃ … … … … 0℃ 99％ 30℃ 32℃ 3℃ 0℃ 100％ 30℃ 32℃ 3℃ 1℃ 0％ 2℃ 3℃ 1℃ 1℃ 1％ 2℃ 3℃ 1℃ 1℃ 2％ 2℃ 3℃ 1℃ 1℃ 3％ 2℃ 3℃ 1℃ 1℃ … … … … 1℃ 99％ 30℃ 32℃ 3℃ 1℃ 100％ 30℃ 32℃ 3℃ … … … … …

[0109] It can be seen from Table 1 that in addition to different ambient temperatures, business pressures, actual temperature values ​​of the components to be measured and the temperature values ​​of the temperature reference points, Table 1 also includes an allowable error range (reference temperature difference). It can be understood that the above error range is determined based on the difference between the actual temperature value of the component to be measured and the temperature value of the temperature reference point. In a scenario with high business pressure, for example, when the business pressure is 90-100%, the allowable error range is greater than the difference between the actual temperature value of the component to be measured and the temperature value of the temperature reference point. For example, when the business pressure is 99%, the difference between the actual temperature value of 32°C of the component to be measured and the temperature value of 30°C of the temperature reference point is 2°C, while the allowable error range is ±3°C. It should be noted that Table 1 is for illustration only and is not intended to limit the embodiments of the present invention.

[0110] In one embodiment, when the normal heat dissipation model is established by the first method, step 102 may be performed as follows:

[0111] S1023. Obtain current business pressure parameters and working environment temperature of the component to be tested.

[0112] S1024. In a lookup table of actual temperature values ​​and reference temperature differences of the component to be tested, search for the actual temperature value and reference temperature difference of the component to be tested corresponding to the combination of the current business pressure parameter, the working environment temperature, and the first temperature value; wherein the actual temperature value searched is the third temperature value.

[0113] For example, the current ambient temperature is 0°C, the business pressure is 1%, and the current temperature information of the temperature reference point is 2°C. From Table 1, we get that the expected temperature value (third temperature value) of the component to be tested is 3°C, and the reference temperature difference is 1°C.

[0114] The second method is to establish a temperature estimation model by establishing a functional relationship, which may include the following steps:

[0115] S1025. Obtain an actual temperature value of the component to be tested within a preset working environment temperature range when it works normally under different business pressures and a reference temperature value of the target component.

[0116] The implementation method in this step is the same as that in step S1021 and will not be repeated here.

[0117] S1026. The actual temperature value of the component to be tested and the reference temperature value of the target component under different ambient temperatures and business pressures are summarized and fitted, and the allowable difference between the actual temperature value and the reference temperature value under different combinations is set as the reference temperature difference, thereby obtaining a functional relationship between the actual temperature value of the component to be tested and the reference temperature difference.

[0118] Among them, the functional relationship between the actual temperature value of the component to be tested and the reference temperature difference can be fitted according to the data in Table 1, and this application will not go into details here.

[0119] In one embodiment, when the normal heat dissipation model is established by the second method, step 102 may be performed as follows:

[0120] S1027. Obtain current business pressure parameters and working environment temperature of the component to be tested.

[0121] S1028. Determine the actual temperature value and the reference temperature difference of the component to be tested according to the functional relationship between the actual temperature value and the reference temperature difference of the component to be tested, the current business pressure parameter, and the working environment temperature; wherein the determined actual temperature value is the third temperature value.

[0122] Whether the temperature estimation model is established by a lookup table in the first method, or the temperature estimation model is established by a functional relationship in the second method, after it is built into the software and the resident thread is started, it is helpful to provide a reliable basic data for abnormal temperature detection of the component to be tested, so as to facilitate normal heat dissipation of the component to be tested.

[0123] S103, based on whether the temperature difference between the second temperature value and the third temperature value is greater than the reference temperature difference, confirm whether the second temperature value is an abnormal temperature value. In one scenario, based on the temperature difference between the second temperature value and the third temperature value being less than or equal to the reference temperature difference, S101 is executed. In another scenario, based on the temperature difference between the second temperature value and the third temperature value being greater than the reference temperature difference, to confirm that the third temperature value is an abnormal temperature value, S104 is executed.

[0124] S104: Generate alarm information and control the fan to perform abnormal speed regulation.

[0125] In this step, after confirming that the third temperature value is an abnormal temperature value, an alarm message is generated and the fan is controlled to perform abnormal speed regulation. For example, when the ambient temperature is 0°C and the business pressure is 99%, the temperature information of the temperature reference point is 30°C. From the table, it can be seen that the expected temperature value of the component to be tested, such as NPU, is 32°C. However, the BMC obtains the self-test temperature value of NPU through the redfish protocol, which is 36°C. At this time, the difference between the expected temperature value of 32°C and the self-test temperature value of 36°C is 4°C, which exceeds the allowable error range of 3°C, then an alarm message is generated and the fan is controlled to perform abnormal speed regulation.

[0126] It is understandable that when the BMC detects an abnormal temperature value, it will send a self-generated alarm message to a specific management console, monitoring system or administrator according to a preset strategy. This alarm message is usually sent through a network management protocol (such as SNMP (simple network management protocol)), IPMI message, system log or email. The specific alarm configuration and recipient can be pre-set in the BMC configuration, and the embodiment of the present application is not limited here.

[0127] Abnormal speed regulation indicates that the fan is running at a constant speed value. When the fan is running at a constant speed value, the component to be tested can operate normally without overheating and card failure. In another possible implementation, when the self-test temperature value of the component to be tested is verified to be an abnormal temperature value, the workload of the component can be reduced to avoid overheating or overloading of the component and protect the component from damage.

[0128] The temperature value verification method provided in the embodiment of the present application realizes the verification of the component temperature value through four steps: selecting the component, selecting the temperature reference point, measuring and fitting the component temperature value, and real-time verification. The server BMC can solve the problem of obtaining an abnormal temperature value from the component through the software protocol, adjusting the speed according to the abnormal temperature value, and causing the component to overheat and fail.

[0129] It is understandable that the size of the sequence number of each step in the above-mentioned embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. In addition, in some possible implementations, the steps in the above-mentioned embodiments can be selectively executed according to actual conditions, and can be partially executed or fully executed, which is not limited here. In addition, all or part of any features of any of the above-mentioned embodiments can be freely and arbitrarily combined without contradiction; the combined technical solutions are also within the scope of the present application.

[0130] like Figure 3 As shown, an embodiment of the present application also provides a temperature value verification device, including: a preprocessing module 101, an acquisition module 102, a temperature determination module 103, a verification module 104 and a processing module 105.

[0131] The pre-processing module 101 is used to select a component with the smallest distance from the component to be tested and provided with a temperature sensor as a target component.

[0132] An acquisition module 102 is used to acquire a first temperature value of a target component and a second temperature value of a component to be measured; wherein the target component and the component to be measured have a correlation, and the correlation indicates that the magnitude of the first temperature value can affect the magnitude of the second temperature value;

[0133] The temperature determination module 102 is used to determine the third temperature value and the reference temperature difference of the component to be tested according to the first temperature value and the pre-constructed temperature estimation model; wherein the temperature estimation model is established according to the temperature values ​​of the target component and the component to be tested when they are working normally under different business pressures within a preset working environment temperature range, and the temperature estimation model is used to indicate the correlation between the first temperature value of the target component, the third temperature value of the component to be tested and the reference temperature difference under different ambient temperatures and business pressures; the ambient temperature indicates the temperature value of the external environment when the component to be tested is running; the business pressure indicates the size of the workload of the component to be tested when the business is running;

[0134] The verification module 103 is used to determine whether the second temperature value is an abnormal temperature value based on whether the temperature difference between the second temperature value and the third temperature value is greater than a reference temperature difference.

[0135] In a possible implementation manner, the preprocessing module 101 is further configured to select a temperature sensor pre-configured on the side of the component to be measured as the target component, and use the pre-configured temperature sensor as a temperature reference point of the component to be measured.

[0136] In one possible implementation, the preprocessing module 101 is also used to pre-build a temperature estimation module, including: determining a component that has the smallest distance from the component to be tested and is equipped with a temperature sensor as a target component; obtaining an actual temperature value of the component to be tested within a preset working environment temperature range when working normally under different business pressures and a reference temperature value of the target component; summarizing and combining the actual temperature values ​​of the component to be tested and the reference temperature values ​​of the target component under different ambient temperatures and business pressures, and setting the allowable difference between the actual temperature value and the reference temperature value under different combinations as the reference temperature difference, thereby forming a lookup table of the actual temperature value and the reference temperature difference of the component to be tested.

[0137] In one possible implementation, the acquisition module 102 is also used to obtain the current business pressure parameters and working environment temperature of the component to be tested; in the actual temperature value and reference temperature difference lookup table of the component to be tested, the actual temperature value and reference temperature difference of the component to be tested corresponding to the combination of the current business pressure parameters, the working environment temperature and the first temperature value are searched; wherein the actual temperature value searched is the third temperature value.

[0138] In one possible implementation, the preprocessing module 101 is also used to pre-build a temperature estimation module, including: determining a component that has the smallest distance from the component to be measured and is equipped with a temperature sensor as a target component; obtaining an actual temperature value of the component to be measured within a preset working environment temperature range when working normally under different business pressures and a reference temperature value of the target component; summarizing and fitting the actual temperature value of the component to be measured and the reference temperature value of the target component under different ambient temperatures and business pressures, and setting the allowable difference between the actual temperature value and the reference temperature value under different combinations as the reference temperature difference, thereby obtaining a functional relationship between the actual temperature value of the component to be measured and the reference temperature difference.

[0139] In one possible implementation, the acquisition module 102 is also used to obtain the current business pressure parameters and the working environment temperature of the component to be tested; determine the actual temperature value and the reference temperature difference of the component to be tested based on the functional relationship between the actual temperature value and the reference temperature difference of the component to be tested and the current business pressure parameters and the working environment temperature; wherein the determined actual temperature value is the third temperature value.

[0140] In a possible implementation, the processing module 105 is configured to generate an alarm message in response to a temperature difference between the second temperature value and the third temperature value being greater than a reference temperature difference.

[0141] In one possible implementation, the processing module 105 is also used to trigger abnormal fan speed regulation when it is confirmed that the second temperature value is an abnormal temperature value; wherein the abnormal speed regulation indicates that the fan is running at a constant speed value, and when the fan is running at a constant speed value, the component to be tested can operate normally.

[0142] It should be understood that the preprocessing module 101, the acquisition module 102, the temperature determination module 103, the verification module 104 and the processing module 105 can all be implemented by software, or can be implemented by hardware. Exemplarily, the implementation of the preprocessing module 101 is introduced below by taking the preprocessing module 101 as an example. Similarly, the implementation of the acquisition module 102, the temperature determination module 103, the verification module 104 and the processing module 105 can refer to the implementation of the preprocessing module 101.

[0143] As an example of a software functional unit, the preprocessing module 101 may include code running on a computing instance. Among them, the computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Further, the above-mentioned computing instance may be one or more. For example, the preprocessing module 101 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code can be distributed in the same region or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code can be distributed in the same AZ or in different AZs, and each AZ includes a data center or multiple data centers with similar geographical locations. Among them, usually a region can include multiple AZs.

[0144] Similarly, multiple hosts / virtual machines / containers used to run the code can be distributed in the same virtual private cloud VPC or in multiple VPCs. Usually, a VPC is set up in a region. For cross-region communication between two VPCs in the same region and between VPCs in different regions, a communication gateway needs to be set up in each VPC to achieve interconnection between VPCs through the communication gateway.

[0145] As an example of a hardware functional unit, the preprocessing module 101 may include at least one computing device, such as a server, etc. Alternatively, the preprocessing module 101 may also be a device implemented using an application specific integrated circuit ASIC, or a programmable logic device PLD, etc. The PLD may be implemented using a CPLD, FPGA, GAL, or any combination thereof.

[0146] The multiple computing devices included in the preprocessing module 101 can be distributed in the same region or in different regions. The multiple computing devices included in the preprocessing module 101 can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in the preprocessing module 101 can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, and GALs.

[0147] In this way, the temperature value verification device provided in the embodiment of the present application has the ability to detect whether the self-test temperature value of the component to be tested is an abnormal temperature value, and generates an alarm message and triggers abnormal fan speed adjustment after detecting the abnormal temperature value, thereby solving the problem that when the server BMC obtains an abnormal temperature value from a compatible component, it performs intelligent speed adjustment according to the abnormal temperature value, thereby causing the card to drop due to overheating.

[0148] Next, based on the above content, the structure of a computing device provided in an embodiment of the present application is exemplarily explained to better understand the usage scenario of the present application. It can be understood that part or all of the content of the computing device can refer to the above description.

[0149] For example, Figure 4 FIG. 1 is a schematic diagram showing a structure of an example of a computing device. Figure 4 As shown, the computing device may include a frame 701 and a power supply 721 and a mainboard 710 installed inside the frame 701 . The power supply 721 is electrically connected to the mainboard 710 to supply power to the components on the mainboard 710 .

[0150] Exemplarily, the mainboard includes a CPU (central processing unit) 711 , a memory 712 , a complex programmable logic device (CPLD) 713 , a baseboard management controller (BMC) 714 , a PCIE slot 715 , a network card 722 , a hard disk 723 , and a fan 724 .

[0151] The processor 711 may include any one or more of a neural processing unit (NPU), a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP) or a digital signal processor (DSP).

[0152] Memory 712 is inserted into a memory slot, which serves as an external cache. Exemplarily, memory 712 may be a random access memory (RAM). By way of example but not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0153] Among them, the PCIE slot 715 is suitable for expanding at least one of a graphics processing unit (GPU) card, a network card, a video capture card, an HBA (Host Bus Adapter) card, a RAID (redundant arrays of independent disks) card, and an SSD (solid state disk or solid state drive), and can also support the expansion of various adapter cards.

[0154] Among them, the complex programmable logic device 713 is a digital integrated circuit whose logic functions are constructed by users according to their own needs. In the embodiment of the present application, the CPLD can receive the control signal sent by the BMC, and output PWM signals with different duty cycles according to the control signal. In the present application, the CPLD is only an exemplary description, and the CPLD can also be replaced by a field programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof. In other implementations, the BMC can directly output PWM signals with different duty cycles.

[0155] The computing device may further include a speed sensor (not shown), which may be disposed near the fan 724 (such as near the bearing or motor of the fan 724) to detect the rotation speed of the fan 724. The fan 724 may be used to reduce the temperature of hardware in the computing device (including but not limited to the frame 702 of the computing device, as well as the power supply 721, the motherboard 710, the CPU 711, the memory 712, the complex programmable logic device 713, the BMC 714, the PCIE slot 715, the network card 722, the hard disk 723, etc.).

[0156] Among them, BMC714 can be used to implement a series of monitoring and control functions for hardware, such as detecting the operating status of the temperature, voltage, etc. of each component of the computing device (CPU711, memory 712, complex programmable logic device 713, PCIE slot 715, network card 722, hard disk 723). Specifically, BMC714 has a BMC interface, and BMC can monitor the temperature information of the corresponding position of each component in the computing device, and according to the obtained temperature information, the fan corresponding to each component in the computing device generates a corresponding fan control signal, and controls the fan 724 to run at a corresponding speed to ensure that each component of the computing device is in a healthy state. BMC714 can also be used to record the information and log records of each hardware of the computing device, which is used to prompt the user and locate subsequent problems. It should be understood that BMC714 can also perform functions other than those described in this article, which are not listed here one by one. It should be noted that in other embodiments, BMC714 has different names in different computing devices, for example, the BMC of Huawei servers and super fusion servers is iBMC, the BMC of HPE servers is called iLO, and the BMC of DELL servers is called iDRAC.

[0157] It should be understood that each component can obtain its own temperature information through a temperature sensor (not shown), which is arranged near or inside the component to ensure that the temperature changes of each component inside the computing device can be monitored more accurately. At the same time, a reference temperature sensor is also arranged around the component to verify the self-test temperature value obtained by each component through its own management software. The reference temperature sensor can be a built-in temperature sensor of the relevant component or a temperature sensor reserved during hardware design. As long as the self-test temperature value of the component can be accurately verified for abnormal temperature values, this application is not limited here. In addition, the temperature sensor can also be set at the air inlet, air outlet, area with high concentration of electronic devices or other areas of the computing device to detect the ambient temperature of various areas inside the computing device.

[0158] Next, taking the network card 722 in the computing device as an example of a component to be tested, the selection of a target component having correlation with the network card 722 is described according to the solution provided in this embodiment. Figure 4As shown, the network card 722 is located at the side of the frame 701, and the component with the shortest distance from the network card 722 and equipped with a temperature sensor is the BMC 714, so the BMC 714 can be used as the target component to verify the temperature value of the network card 722. Optionally, in order to make the temperature value verification of the network card 722 more accurate, the developer can also install a reference temperature sensor 714 on the side of the network card 722, and use the reference temperature sensor 714 as a temperature reference point to verify the temperature value of the network card 722. It should be noted that when the reference temperature sensor 714 is considered to be reserved during hardware design, it needs to be as close to the component to be tested as possible.

[0159] In some embodiments, if the component to be tested is a GPU module plugged into the PCIE slot 715, the target component can select the complex programmable logic device 713, which is the component with the smallest distance from the GPU module and is provided with a temperature sensor. Alternatively, if the processor 711 is a neural processing unit (NPU), the component to be tested is an NPU, and the target component can be a complex programmable logic device 713. Among them, the hard disk 723 is divided into a mechanical hard disk (hard disk drive, HDD) and a solid state drive (Solid State Disk or Solid State Drive, SSD). Exemplarily, the solid state drive can be a solid state drive of the PCIe channel, such as a solid state drive of the logical device interface specification (nonvolatile memory express, NVME, used to access non-volatile memory media attached through the PCI Express (PCIe) bus). It should be understood that the hard disk 723 is only an example of a non-volatile memory and does not constitute a specific limitation. In practical applications, a non-volatile memory can be selected in combination with actual conditions.

[0160] It should be noted that Figure 4 It is only an example of a computing device and does not constitute a specific limitation. In practical applications, it may include Figure 4More or fewer devices, for example, may also include a single-chip microcomputer (an integrated circuit chip, equivalent to a microcomputer), a digital signal processor (DSP), an application specific integrated circuit (ASIC), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, general-purpose processors, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. It should be noted that the system architecture and application scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. It is known to those skilled in the art that with the evolution of the system architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0161] The present application also provides a computing device in an embodiment. The computing device includes a baseboard management controller. The baseboard management controller can execute the following steps: Figure 2 And the technical scheme corresponding to the above protection makes the terminal device have the technical effect of the technical scheme corresponding to the above protection.

[0162] The present application also provides a computer-readable storage medium. The computer-readable storage medium is used to store computer program instructions. When the computer program instructions are executed on a computing device, the computing device executes the temperature value verification method involved in the above embodiment. The computer-readable storage medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0163] The embodiment of the present application also provides a computer program product including instructions. When the computer program product is run on a computing device, the computing device is enabled to execute the temperature value verification method involved in the above embodiment.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application. Those skilled in the art should understand that, although the present application is described in detail with reference to the aforementioned embodiments, the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions in the embodiments of the present application.

Claims

1. A temperature value verification method, characterized in that: include: Acquire a first temperature value of a target component and a second temperature value of a component to be measured; wherein the target component and the component to be measured have a correlation, and the correlation indicates that the magnitude of the first temperature value can affect the magnitude of the second temperature value; Determine the third temperature value and the reference temperature difference of the component to be tested according to the first temperature value and a pre-built temperature estimation model; the temperature estimation model is used to indicate the correlation between the first temperature value of the target component, the third temperature value of the component to be tested and the reference temperature difference under different ambient temperatures and business pressures; the ambient temperature indicates the temperature value of the external environment when the component to be tested is running; the business pressure indicates the size of the workload of the component to be tested when carrying business operations; Based on the temperature difference between the second temperature value and the third temperature value and the reference temperature difference, it is determined whether the second temperature value is an abnormal temperature value.

2. The method according to claim 1, characterized in that Also includes: The component with the smallest distance from the component to be tested and equipped with a temperature sensor is selected as the target component.

3. The method according to claim 1 or 2, characterized in that: The temperature estimation model includes an actual temperature value of the component to be measured and a reference temperature difference lookup table, and the method further includes: Acquire the actual temperature value of the component to be tested when it works normally under different business pressures within a preset working environment temperature range and the reference temperature value of the target component; The actual temperature value of the component to be tested and the reference temperature value of the target component under different ambient temperatures and business pressures are summarized and combined, and the allowable difference between the actual temperature value and the reference temperature value under different combinations is set as the reference temperature difference to obtain a lookup table of the actual temperature value and reference temperature difference of the component to be tested.

4. The method according to claim 3, characterized in that The step of determining the third temperature value and the reference temperature difference of the component to be tested according to the first temperature value and a pre-built temperature estimation model comprises: Obtain the current business pressure parameters and working environment temperature of the component to be tested; In the actual temperature value and reference temperature difference lookup table of the component to be tested, the actual temperature value and reference temperature difference of the component to be tested corresponding to the combination of the current business pressure parameter, the working environment temperature and the first temperature value are searched; wherein the actual temperature value searched is the third temperature value.

5. The method according to any one of claims 1 to 4, characterized in that: Also includes: In response to a temperature difference between the second temperature value and the third temperature value being greater than the reference temperature difference, an alarm message is generated.

6. The method according to claim 5, characterized in that Also includes: When it is confirmed that the second temperature value is an abnormal temperature value, the fan is controlled to perform abnormal speed regulation.

7. The method according to any one of claims 1 to 6, characterized in that: The components to be tested include one or more of a smart network card, a network card optical module, an NPU module and a GPU module.

8. The method according to claim 7, characterized in that The component to be tested is an intelligent network card, and the target component is a baseboard management controller.

9. The method according to claim 7, characterized in that: The component to be tested is an intelligent network card, and the target component is a pre-configured reference temperature sensor.

10. A computing device, characterized in that: The computing device comprises a baseboard management controller, and the baseboard management controller is used to execute the method according to any one of claims 1-9.

Images