Method and system for monitoring running state of BMC (Baseboard Management Controller) of microcomputer host

Abstract

The invention relates to the technical field of anomaly monitoring, in particular to a BMC running state monitoring method and system for a microcomputer host, and the method comprises the steps: determining the temperature anomaly degree according to the abnormal change of the CPU temperature after the rotation speed of a radiator influences the CPU temperature and the abnormal rising trend of the CPU temperature; determining the abnormal heat dissipation degree according to the characteristic that the temperature of the CPU is usually reduced after the rotating speed of the radiator is increased; on the basis of the two, the alarm triggering risk is determined by combining the stable condition of the host operation state; determining an alarm response demand degree according to the numerical value abnormity of the CPU temperature at all moments from the occurrence of the overtemperature to the alarm; according to the method, the more accurate dynamic de-jitter time window is obtained by combining the window adjustment coefficient adaptively determined by the alarm triggering risk and the alarm response demand degree, so that the accuracy of alarm detection is higher, and the effect of monitoring the BMC running state of the microcomputer host according to the dynamic de-jitter time window is better.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormality monitoring, and in particular to a method and system for monitoring the BMC operating status of a microcomputer host. Background Art

[0002] With the continuous development of information technology, microcomputer hosts are increasingly used in various information systems, from personal computers to enterprise-level servers to edge computing devices. Microcomputer hosts have become indispensable and important equipment in various computing scenarios due to their small size, strong performance, and high energy efficiency. In microcomputer hosts, the baseboard management controller (BMC) is a key management component, which is mainly responsible for hardware monitoring, diagnosis, fault detection, and remote management of the host. Therefore, real-time monitoring of the operating status of the BMC can help system administrators discover potential problems in a timely manner, perform fault diagnosis and preventive maintenance, and ensure the normal operation of the system.

[0003] Considering that the temperature detected by the temperature sensor has a certain error, the alarm mechanism in the existing BMC usually adds a de-jitter function based on temperature monitoring to avoid false alarms, so as to perform more accurate BMC operation status monitoring of the microcomputer host. However, the existing technology usually uses a fixed de-jitter time window for alarm detection based on the CPU temperature, but the setting of the de-jitter time window also means that there is a time delay in the alarm. If the microcomputer host is seriously abnormal, then when the fixed de-jitter time window is used for alarm detection, the system may fail or be damaged due to the lack of timely alarm; on the contrary, if the microcomputer host is slightly abnormal or there is no abnormality, then when the fixed de-jitter time window is used for alarm detection, false alarms may occur; therefore, the accuracy of the existing technology using a fixed de-jitter time window for alarm detection based on the CPU temperature is low, resulting in poor monitoring of the BMC operation status of the microcomputer host. Summary of the invention

[0004] In order to solve the problem of low accuracy of alarm detection using a fixed de-jitter time window according to CPU temperature in the prior art, the present application aims to provide a BMC operation status monitoring method and system for a microcomputer host. The technical solutions adopted are as follows: The first aspect of the present application provides a method for monitoring the BMC operation status of a microcomputer host, comprising: During the operation of the microcomputer host, the prior alarm time, CPU temperature curve, radiator speed curve and CPU utilization curve of each monitoring time period monitored by the BMC are collected; the preset de-jitter time window and the CPU over-temperature start time when the CPU temperature exceeds the preset temperature threshold for the first time in each monitoring time period are obtained; Determine the degree of temperature anomaly based on the change correlation between the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve; determine the degree of heat dissipation anomaly based on the matching of the CPU temperature change during the radiator speed increase process; determine the alarm risk of each monitoring time period based on the overall stability of the CPU utilization curve, the temperature anomaly degree and the heat dissipation anomaly degree; In each monitoring time period, the alarm response requirement level of each monitoring time period is determined according to the time length between the prior alarm moment and the CPU overtemperature start moment, the CPU temperature distribution and the overall CPU temperature; Determine the window adjustment coefficient of each monitoring time period according to the alarm response requirement degree and the alarm risk; adjust the preset de-jitter time window according to the window adjustment coefficient to determine the dynamic de-jitter time window of each monitoring time period; monitor the BMC operation status of the microcomputer host according to the dynamic de-jitter time window.

[0005] Furthermore, the process of obtaining the temperature anomaly degree includes: On the CPU temperature curve, the slope value of the line between the data point corresponding to each sampling moment and the data point corresponding to the previous sampling moment is used as the temperature slope value at each sampling moment; and the temperature rising trend characteristic value is determined according to the average value of the temperature slope values ​​at all sampling moments; On the CPU temperature curve, negative correlation mapping is performed between the mean value of the CPU temperature at all sampling moments and the difference between the preset temperature threshold value to determine the proximity degree of the threshold temperature; Determine the degree of correlation between the temperature and the rotation speed according to a negative correlation mapping of the Pearson correlation coefficient between the CPU temperature curve and the heat sink rotation speed curve; The degree of temperature anomaly in each monitoring time period is determined according to the temperature-speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value; the temperature-speed correlation degree is negatively correlated with the temperature anomaly degree; the threshold temperature proximity degree and the temperature rising trend characteristic value are both positively correlated with the temperature anomaly degree.

[0006] Furthermore, the process of determining the degree of temperature anomaly in each monitoring time period according to the temperature-rotation speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value includes: The product of the negative correlation mapping value of the temperature-rotation speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value is normalized to determine the temperature anomaly degree of each monitoring time period.

[0007] Furthermore, the process of obtaining the abnormal degree of heat dissipation includes: On the radiator speed curve, the slope value of the line between the data point corresponding to each sampling moment and the data point corresponding to the previous sampling moment is used as the speed slope value at each sampling moment; the sampling moment when the speed slope value is greater than 0 is used as the speed rising moment; The difference between the speed slope value and the corresponding temperature slope value at each speed rising moment is negatively correlated to determine the instantaneous heat dissipation fault degree at each speed rising moment; the heat dissipation abnormality degree of each monitoring time period is determined based on the average of the instantaneous heat dissipation fault degrees at all speed rising moments.

[0008] Furthermore, the process of obtaining the alarm risk includes: The CPU utilization standard deviations at all sampling moments on the CPU utilization curve are negatively correlated to determine the stability of the CPU utilization in each monitoring time period; The product of the negative correlation mapping value of the CPU utilization stability, the temperature abnormality and the heat dissipation abnormality is normalized to determine the alarm risk of each monitoring time period.

[0009] Furthermore, the process of obtaining the alarm response requirement degree includes: The monitoring time period in which both the priori alarm moment and the CPU overtemperature start moment exist and the CPU overtemperature start moment is before the priori alarm moment is used as the analysis time period; the alarm response requirement degree of other monitoring time periods outside the analysis time period is set to the preset requirement degree; In each analysis time period, all sampling moments between the CPU over-temperature start moment and the prior alarm moment are taken as delay moments; the delay moment when the corresponding CPU temperature is greater than the preset temperature threshold is taken as the reference abnormal moment; the difference between the CPU temperature at each reference abnormal moment and the preset temperature threshold is taken as the instantaneous over-temperature degree at each reference abnormal moment; the overall over-temperature degree of each analysis time period is determined based on the accumulated value of the instantaneous over-temperature degrees at all reference abnormal moments; The product between the number of delay moments and the overall over-temperature degree in each analysis time period is normalized to determine the alarm response requirement degree in each analysis time period.

[0010] Furthermore, the process of obtaining the window adjustment coefficient includes: A negative correlation mapping is performed on the product of the alarm response requirement degree and the alarm risk to determine the window adjustment coefficient of each monitoring time period.

[0011] Furthermore, the process of acquiring the dynamic de-jittering time window includes: The dynamic de-jitter time window of each monitoring time period is determined by multiplying the product of the window adjustment coefficient and the preset de-jitter time window.

[0012] Furthermore, the process of monitoring the BMC operating status of the microcomputer host according to the dynamic de-jittering time window includes: The dynamic de-jitter time window of each monitoring time period is used as the final de-jitter time window of each subsequent monitoring time period until a new dynamic de-jitter time window is re-determined when an alarm occurs; the BMC operating status of the microcomputer host is monitored based on the final de-jitter time window of each monitoring time period combined with the corresponding CPU temperature.

[0013] In a second aspect, the present application provides a BMC operation status monitoring system for a microcomputer host, the system comprising: The data collection preprocessing module is used to collect the prior alarm time, CPU temperature curve, radiator speed curve and CPU utilization curve of each monitoring time period monitored by the BMC during the operation of the microcomputer host; obtain the preset de-jitter time window and the CPU over-temperature start time when the CPU temperature exceeds the preset temperature threshold for the first time in each monitoring time period; The first determination module is used to determine the temperature abnormality according to the change correlation between the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve; determine the heat dissipation abnormality according to the matching of the CPU temperature change during the radiator speed increase process; determine the alarm risk of each monitoring time period according to the overall stability of the CPU utilization curve, the temperature abnormality and the heat dissipation abnormality; A second determination module is used to determine the alarm response requirement degree of each monitoring time period according to the time length between the prior alarm moment and the temperature abnormality moment, the CPU temperature distribution and the overall CPU temperature in each monitoring time period; An operation status monitoring module is used to determine the window adjustment coefficient of each monitoring time period according to the alarm response requirement degree and the alarm risk; adjust the preset de-jitter time window according to the window adjustment coefficient to determine the dynamic de-jitter time window of each monitoring time period; and monitor the BMC operation status of the microcomputer host according to the dynamic de-jitter time window.

[0014] In a third aspect, the present application provides a computer device, comprising a memory and a processor. The memory is used to store computer program code, and the processor is used to call and run the computer program code from the memory to execute the method of the first aspect or any embodiment of the first aspect of the present application.

[0015] In a fourth aspect, the present application provides a computer program product, comprising a computer program code, which, when executed, performs the method of the first aspect of the present application or any embodiment of the first aspect.

[0016] In a fifth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program code. When the computer program code is executed, the method of the first aspect or any embodiment of the first aspect of the present application is performed.

[0017] This application has the following beneficial effects: The present application first determines the degree of temperature abnormality according to the abnormal change of CPU temperature after the radiator speed affects the CPU temperature and the abnormal rising trend of CPU temperature; then determines the degree of heat dissipation abnormality according to the characteristic that the CPU temperature usually decreases after the radiator speed increases; then determines the risk of triggering an alarm based on the degree of temperature abnormality and the degree of heat dissipation abnormality in combination with the CPU utilization curve that characterizes the stability of the host operation state; then determines the degree of alarm response requirement according to the numerical abnormality of the CPU temperature between the start time of CPU overheating and the alarm time; thereby adaptively determining a more accurate window adjustment coefficient for adjusting the de-jitter time window in combination with the alarm risk and the degree of alarm response requirement; finally, adaptively determining a more accurate dynamic de-jitter time window according to the window adjustment coefficient, so that the accuracy of alarm detection according to the dynamic de-jitter time window is higher, which solves the problem of low accuracy of alarm detection according to the CPU temperature using a fixed de-jitter time window in the prior art, and also makes the effect of BMC operation status monitoring of the microcomputer host according to the dynamic de-jitter time window better. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a method for monitoring the BMC operating status of a microcomputer host provided by an embodiment of the present invention; Figure 2 A structural diagram of a BMC operation status monitoring system for a microcomputer host provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the BMC operation status monitoring method and system of a microcomputer host proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0021] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0022] The following is a detailed description of a method and system for monitoring the BMC operating status of a microcomputer host provided by the present invention in conjunction with the accompanying drawings.

[0023] The present application embodiment provides a method for monitoring the BMC operation status of a microcomputer host. Figure 1 , which shows a flow chart of a BMC operating status monitoring method for a microcomputer host provided by an embodiment of the present invention, the method comprising: Step S101: During the operation of the microcomputer host, the prior alarm time, CPU temperature curve, radiator speed curve and CPU utilization curve of each monitoring time period monitored by the BMC are collected; the preset de-jitter time window and the CPU over-temperature start time when the CPU temperature first exceeds the preset temperature threshold in each monitoring time period are obtained.

[0024] In a specific implementation of an embodiment of the present invention, the CPU temperature at each sampling moment is collected by a temperature sensor, and the fan speed of the radiator and the CPU utilization rate at each sampling moment are collected by a hardware monitoring system of a microcomputer host; and the CPU temperature, the fan speed of the radiator and the CPU utilization rate are transmitted to the monitoring module of the BMC to complete the collection; then, all CPU temperatures in each monitoring time period are arranged in time series and curve fitting is performed to obtain a CPU temperature curve; the fan speeds of all radiators in each monitoring time period are normalized and arranged in time series and curve fitting is performed to obtain a radiator speed curve; all CPU utilization rates in each monitoring time period are arranged in time series and curve fitting is performed to obtain a CPU utilization curve; wherein, the time length of each monitoring time period is set to 1 minute, the sampling frequency is set to collect once per second, and the time length of the preset de-jitter time window is set to 5 seconds, which can be adjusted according to the specific implementation environment.

[0025] In a specific implementation of an embodiment of the present invention, the preset temperature threshold is set to 90 degrees Celsius and can be adjusted at will; the prior alarm moment is the moment corresponding to the triggering alarm mechanism monitored in each monitoring time period; for each monitoring time period, if the CPU temperature is continuously greater than the preset temperature threshold for a period of time greater than the corresponding de-jitter time window, the alarm mechanism will be triggered.

[0026] Step S102: Determine the degree of temperature abnormality based on the correlation between the changes in the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve; determine the degree of heat dissipation abnormality based on the matching of the CPU temperature changes during the radiator speed increase process; determine the alarm risk of each monitoring time period based on the overall stability of the CPU utilization curve, the degree of temperature abnormality and the degree of heat dissipation abnormality.

[0027] First, the problem that needs to be solved is that the accuracy of alarm detection based on CPU temperature using a fixed de-jitter time window is low. The reason is that the setting of the de-jitter time window also means that there is a time delay in the alarm. If the microcomputer host is seriously abnormal, then when the fixed de-jitter time window is used for alarm detection, the system may fail or be damaged due to the lack of timely alarm. On the contrary, if the microcomputer host is slightly abnormal or has no abnormality, then when the fixed de-jitter time window is used for alarm detection, false alarms may occur. Therefore, it is necessary to adaptively adjust the de-jitter time window in combination with the abnormal state of the microcomputer host.

[0028] When the microcomputer host is running abnormally, the temperature usually changes abnormally; therefore, in terms of temperature analysis, first of all, the more obvious the CPU temperature rise trend is, the more likely it is to trigger the alarm mechanism, and the temperature is closer to abnormality; and the higher the radiator speed is, the stronger the ability to suppress the CPU temperature is. If the radiator fan speed adjustment follows the change in CPU temperature, that is, the correlation between the CPU temperature and the radiator speed is high, the CPU temperature will usually decrease with the increase in the radiator speed, resulting in a lower possibility of triggering the warning mechanism; therefore, the degree of temperature abnormality is first determined based on the change correlation between the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve, so that the greater the temperature abnormality, the higher the risk of triggering the alarm.

[0029] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the degree of temperature anomaly includes: On the CPU temperature curve, the slope value of the line between the data point corresponding to each sampling moment and the data point corresponding to the previous sampling moment is used as the temperature slope value at each sampling moment; the temperature rising trend characteristic value is determined according to the mean of the temperature slope values ​​at all sampling moments; on the CPU temperature curve, the difference between the mean of the CPU temperature at all sampling moments and the preset temperature threshold is negatively correlated to determine the degree of proximity to the threshold temperature; the degree of temperature-speed correlation is determined according to the negative correlation mapping of the Pearson correlation coefficient between the CPU temperature curve and the radiator speed curve. In a specific implementation of an embodiment of the present invention, the preset temperature threshold is set to 90 degrees, which can be adjusted according to the specific implementation environment.

[0030] According to the process of obtaining the temperature slope value, it can be known that the larger the mean value of the temperature slope values ​​at all sampling moments, that is, the larger the characteristic value of the temperature rising trend, the more obvious the overall rising trend of the CPU temperature curve is, and the more it has the characteristics of the trend of triggering the alarm mechanism; and when the threshold temperature is closer, the CPU temperature at each sampling moment is closer to the preset temperature threshold, which means that the CPU temperature is closer to the alarm value, the corresponding alarm trend is more obvious, and the possibility of triggering the alarm mechanism is higher; and when the temperature-speed correlation degree obtained by the negative correlation mapping of the Pearson correlation coefficient of the CPU temperature curve and the radiator speed curve is greater, it means that the CPU temperature and the radiator speed are more negatively correlated, and the more it conforms to the characteristic that the CPU temperature usually decreases with the increase of the radiator speed, the corresponding correlation is greater, and the possibility of triggering the warning mechanism is lower.

[0031] Therefore, the degree of temperature anomaly in each monitoring time period is further determined according to the temperature-speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value; the temperature-speed correlation degree is negatively correlated with the temperature anomaly degree; the threshold temperature proximity degree and the temperature rising trend characteristic value are both positively correlated with the temperature anomaly degree. Preferably, the process of determining the degree of temperature anomaly in each monitoring time period according to the temperature-speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value includes: normalizing the product of the negative correlation mapping value of the temperature-speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value, and determining the degree of temperature anomaly in each monitoring time period.

[0032] In a specific implementation of the embodiment of the present invention, the process of obtaining the temperature anomaly degree is expressed by the formula: ;in, For the The degree of temperature anomaly during the monitoring period; For the The number of sampling moments in a monitoring period; For the The monitoring period The temperature slope value at each sampling moment; For the Temperature rising trend characteristic value for each monitoring period; For the Pearson correlation coefficient between CPU temperature curve and radiator speed curve in each monitoring period; For the The degree of correlation between temperature and speed in each monitoring period; For the The average CPU temperature at all sampling times in a monitoring period; is the preset temperature threshold; For the The proximity of the threshold temperature for each monitoring period; is the absolute value symbol. For the The negative correlation mapping value of the temperature-speed correlation degree in each monitoring time period; is an exponential function with a natural constant as base; It is a linear normalization function. By normalizing to between 0 and 1, it prevents the temperature anomaly from taking negative values ​​and affecting subsequent calculations.

[0033] Then, for the radiator, under normal circumstances, when the radiator speed increases, it will blow out faster wind, and the rising trend of temperature will be suppressed or even show a downward trend; if the temperature still maintains an upward trend when the radiator speed increases, it means that there are abnormalities or failures in radiator-related components such as heat pipes and silicone grease, and the higher the abnormality of the radiator, the higher the risk of alarm; therefore, the degree of heat dissipation abnormality is determined according to the matching of CPU temperature changes during the process of radiator speed increase. In addition, the stability of CPU utilization can reflect the operating status of the microcomputer host to a certain extent. Therefore, the alarm risk is determined based on the temperature abnormality and heat dissipation abnormality combined with the overall stability of the CPU utilization curve, so that the greater the alarm risk, the more abnormal the microcomputer host is, and the corresponding de-jitter time window should be smaller.

[0034] Preferably, in a specific implementation of the embodiment of the present invention, the process of obtaining the degree of heat dissipation abnormality includes: On the radiator speed curve, the slope value of the line connecting the data point corresponding to each sampling moment and the data point corresponding to the previous sampling moment is used as the speed slope value at each sampling moment; the sampling moment when the speed slope value is greater than 0 is used as the speed rising moment; the difference between the speed slope value and the corresponding temperature slope value at each speed rising moment is negatively correlated to determine the instantaneous heat dissipation fault degree at each speed rising moment; the heat dissipation abnormality degree of each monitoring time period is determined according to the average of the instantaneous heat dissipation fault degrees at all speed rising moments. At each speed rising moment, the more the temperature shows a downward trend, the more normal the radiator state is; therefore, the smaller the instantaneous heat dissipation fault degree obtained by negatively correlating the difference between the speed slope value and the corresponding temperature slope value, the more normal the radiator state at the corresponding sampling moment is; further, the heat dissipation abnormality degree characterizing the radiator abnormality is determined in combination with all sampling moments.

[0035] In a specific implementation of the embodiment of the present invention, the process of obtaining the abnormal degree of heat dissipation is expressed by the formula: ;in, For the The degree of heat dissipation abnormality during each monitoring period; For the The number of speed increase moments in a monitoring time period; For the In the monitoring period The speed slope value at the moment of speed rise; For the In the monitoring period The temperature slope value at each speed increase moment; For the In the monitoring period The instantaneous heat dissipation failure degree at the moment of speed increase; is an exponential function with a natural constant as its base.

[0036] Preferably, in a specific implementation of the embodiment of the present invention, the process of obtaining the alarm risk includes: The CPU utilization standard deviation of all sampling moments on the CPU utilization curve is negatively correlated and mapped to determine the stability of CPU utilization in each monitoring time period. The larger the CPU utilization standard deviation, that is, the smaller the CPU utilization stability, the more unstable the CPU utilization, the more abnormal the corresponding microcomputer host's operating state, the greater the possibility of abnormality, and the higher the risk of alarm. Since the greater the abnormality of the radiator and the greater the temperature abnormality, the smaller the CPU utilization stability, the higher the risk of alarm, the negative correlation mapping value of the CPU utilization stability, the temperature abnormality, and the heat dissipation abnormality are further normalized to determine the risk of alarm in each monitoring time period, so that when the risk of alarm is greater, the microcomputer host is more abnormal, and the corresponding de-jitter time window should be smaller.

[0037] In a specific implementation of the embodiment of the present invention, the process of obtaining the alarm risk is expressed by the formula: ;in, For the The risk of triggering an alarm during each monitoring period; For the The standard deviation of CPU utilization at all sampling times on the CPU utilization curve of a monitoring period; is an exponential function with a natural constant as base; For the The stability of CPU utilization during each monitoring period; For the The degree of heat dissipation abnormality during each monitoring period; For the The degree of temperature anomaly during the monitoring period; is a linear normalization function.

[0038] Step S103: In each monitoring time period, the alarm response requirement level of each monitoring time period is determined according to the time length between the prior alarm time and the CPU overtemperature start time, the CPU temperature distribution and the overall CPU temperature.

[0039] It is further necessary to consider that the purpose of setting the de-jitter time window, that is, the purpose of setting the delay, is to reduce the false alarm rate of the alarm mechanism. However, if the delay is too high, the alarm mechanism will not respond to real abnormalities quickly enough, so it is necessary to shorten the delay, that is, shorten the de-jitter time window; considering that blindly shortening the delay will lead to an increase in the false alarm rate of the alarm, in order to adjust the de-jitter time window more reasonably and scientifically, it is further necessary to adjust it in combination with the alarm response requirements of each monitoring time period. For the monitoring time period in which there are both the prior alarm moment at which an alarm has been issued and the CPU overtemperature starting moment, if the time length between the prior alarm moment and the temperature abnormality moment is long, it means that the time delay between the temperature exceeding the threshold and the alarm behavior is long, which to a certain extent means that the time lag of the alarm mechanism of this monitoring time period is high, and the corresponding response delay should be reduced; and between the prior alarm moment and the CPU overtemperature starting moment, the more times the preset temperature threshold is exceeded and the higher the exceeded temperature is, the more it is necessary to trigger the alarm, and the corresponding time delay should be smaller; therefore, the time length between the prior alarm moment and the temperature abnormality moment, the CPU temperature distribution and the overall CPU temperature are comprehensively considered to determine the degree of alarm response demand; so that the greater the alarm response demand, the smaller the corresponding de-jitter time window should be.

[0040] Preferably, in a specific implementation of the embodiment of the present invention, the process of obtaining the alarm response requirement degree includes: The monitoring time period in which both the priori alarm moment and the CPU overtemperature start moment exist and the CPU overtemperature start moment is before the priori alarm moment is used as the analysis time period; the alarm response requirement degree of other monitoring time periods outside the analysis time period is set to the preset requirement degree. In a specific implementation of the embodiment of the present invention, the preset requirement degree is set to 1, which can be adjusted according to the specific implementation environment; because the calculation of the alarm response requirement degree needs to rely on the time delay from the overtemperature moment to the alarm moment, other situations that do not meet this condition cannot be applied to subsequent analysis, so for the completeness of the embodiment, the alarm response requirement degree in other situations is set to a preset value.

[0041] In each analysis time period, all sampling moments between the CPU overtemperature start moment and the prior alarm moment are taken as delay moments; the corresponding delay moment when the CPU temperature is greater than the preset temperature threshold is taken as the reference abnormal moment; the difference between the CPU temperature at each reference abnormal moment and the preset temperature threshold is taken as the instantaneous overtemperature degree at each reference abnormal moment; the overall overtemperature degree of each analysis time period is determined according to the cumulative value of the instantaneous overtemperature degree of all reference abnormal moments. First, the more the delay moments, the higher the delay, the slower the real abnormal response speed of the corresponding alarm mechanism, and the higher the need to shorten the de-bouncing time window; and the more the reference abnormal moments, the more frequent the instantaneous overtemperature occurs or the longer the overtemperature lasts, the accumulated thermal stress will further affect the service life of the hardware, and the higher the need to shorten the de-bouncing time window should be; the greater the overall overtemperature degree, the greater the overall excess of the reasonable CPU temperature, that is, the preset temperature threshold, and a large excess of the reasonable CPU temperature may cause permanent damage or failure; therefore, the greater the overall overtemperature degree, the higher the need to shorten the de-bouncing time window should be. Therefore, the product between the number of delay moments in each analysis time period and the overall over-temperature degree is further normalized to determine the alarm response requirement degree in each analysis time period.

[0042] In a specific implementation of the embodiment of the present invention, the process of obtaining the alarm response requirement degree is expressed by the formula: ;in, For the The degree of alarm response demand during the analysis period; For the The number of delayed moments in the analysis period; For the The number of reference anomaly moments in the analysis period; For the In the analysis period The CPU temperature at the time of the reference anomaly; is the preset temperature threshold; For the In the analysis period The instantaneous over-temperature degree at the reference abnormal moment; For the The overall degree of over-temperature during the analysis period; is a linear normalization function.

[0043] Step S104: Determine the window adjustment coefficient for each monitoring time period according to the alarm response requirement and the alarm risk; adjust the preset de-jitter time window according to the window adjustment coefficient to determine the dynamic de-jitter time window for each monitoring time period; monitor the BMC operating status of the microcomputer host according to the dynamic de-jitter time window.

[0044] Finally, the alarm risk and the degree of alarm response demand are combined to jointly determine the window adjustment coefficient that characterizes the de-jitter time window adjustment, and the dynamic de-jitter time window is adaptively determined based on the window adjustment coefficient, so that the accuracy of alarm detection based on the dynamic de-jitter time window is higher, that is, the effect of BMC operating status monitoring of the microcomputer host based on the dynamic de-jitter time window is improved.

[0045] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the window adjustment coefficient includes: Since the greater the alarm response requirement, the greater the alarm risk, the smaller the corresponding de-jitter time window should be; therefore, the product between the alarm response requirement and the alarm risk is further negatively correlated to determine the window adjustment coefficient for each monitoring time period. In a specific implementation of an embodiment of the present invention, the process of obtaining the window adjustment coefficient is expressed by the formula: ;in, For the Window adjustment coefficient for each monitoring period; For the The degree of alarm response demand during each monitoring period; For the The risk of triggering an alarm during each monitoring period; It is a preset negative correlation adjustment constant, whose value is greater than or equal to 1. In the embodiment of the present invention, it is set to 1.3, so that the obtained window adjustment coefficient can have the ability to increase and decrease the de-jitter time window, and the preset negative correlation adjustment constant can be adjusted automatically.

[0046] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the dynamic de-jitter time window includes: determining the dynamic de-jitter time window of each monitoring time period by multiplying the product between the window adjustment coefficient and the preset de-jitter time window; which is expressed in the formula as follows: ;in, For the Dynamic de-jitter time window for each monitoring period; It is the preset de-jitter time window; For the It should be noted that each de-jitter time window involved in the calculation represents a corresponding time length by default, because the de-jitter time window is essentially a time length for de-jitter detection.

[0047] Preferably, in some possible implementation methods of the embodiments of the present invention, the process of monitoring the BMC operating status of the microcomputer host according to the dynamic de-jitter time window includes: using the dynamic de-jitter time window of each monitoring time period as the final de-jitter time window of each subsequent monitoring time period until a new dynamic de-jitter time window is re-determined when an alarm occurs; monitoring the BMC operating status of the microcomputer host according to the final de-jitter time window of each monitoring time period in combination with the corresponding CPU temperature. Because the corresponding dynamic de-jitter time window can only be calculated after the end of each monitoring time period, the dynamic de-jitter time window of each monitoring time period is used as the final de-jitter time window of each subsequent monitoring time period of the monitoring time period, and the de-jitter time window is dynamically updated until an alarm occurs. It should be noted that the application of the de-jitter time window in the present application is specifically: an alarm message is issued only when the CPU temperature exceeds the preset temperature threshold for a duration greater than the time length of the corresponding de-jitter time window.

[0048] In summary, the BMC operation status monitoring method of a microcomputer host proposed in the present application first determines the degree of temperature abnormality according to the abnormal change of CPU temperature after the radiator speed affects the CPU temperature and the abnormal rising trend of CPU temperature; then determines the degree of heat dissipation abnormality according to the characteristic that the CPU temperature usually decreases after the radiator speed increases; then determines the alarm risk based on the temperature abnormality and the degree of heat dissipation abnormality combined with the CPU utilization curve that characterizes the stability of the host operation status; then determines the alarm response requirement degree according to the numerical abnormality of the CPU temperature between the start time of CPU overheating and the alarm time; thereby adaptively determining a more accurate window adjustment coefficient for adjusting the de-jitter time window in combination with the alarm risk and the alarm response requirement degree; finally, adaptively determining a more accurate dynamic de-jitter time window according to the window adjustment coefficient, so that the accuracy of alarm detection according to the dynamic de-jitter time window is higher, which solves the problem of low accuracy of alarm detection according to the CPU temperature in the prior art using a fixed de-jitter time window, that is, the effect of BMC operation status monitoring of the microcomputer host according to the dynamic de-jitter time window is better.

[0049] This application also provides a BMC operating status monitoring system for a microcomputer host, see Figure 2, which shows a structural diagram of a BMC operation status monitoring system of a microcomputer host provided by an embodiment of the present invention, the system includes: a data acquisition preprocessing module 201, a first determination module 202, a second determination module 203 and an operation status monitoring module 204.

[0050] The data collection and preprocessing module 201 is used to collect the prior alarm time, CPU temperature curve, radiator speed curve and CPU utilization curve of each monitoring time period monitored by the BMC during the operation of the microcomputer host; obtain the preset de-jitter time window and the CPU over-temperature start time when the CPU temperature exceeds the preset temperature threshold for the first time in each monitoring time period; The first determination module 202 is used to determine the temperature abnormality according to the change correlation between the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve; determine the heat dissipation abnormality according to the matching of the CPU temperature change during the radiator speed increase process; determine the alarm risk of each monitoring time period according to the overall stability of the CPU utilization curve, the temperature abnormality and the heat dissipation abnormality; The second determination module 203 is used to determine the alarm response requirement level of each monitoring time period according to the time length between the prior alarm time and the CPU overtemperature start time, the CPU temperature distribution and the overall CPU temperature in each monitoring time period; The operation status monitoring module 204 is used to determine the window adjustment coefficient of each monitoring time period according to the alarm response requirement degree and the alarm risk; adjust the preset de-jitter time window according to the window adjustment coefficient to determine the dynamic de-jitter time window of each monitoring time period; and monitor the BMC operation status of the microcomputer host according to the dynamic de-jitter time window.

[0051] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the BMC operation status monitoring system of a microcomputer host provided in the above embodiment and the BMC operation status monitoring method embodiment of a microcomputer host belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0052] The present application also provides a computer device. Figure 3, which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present invention, the computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any one of the BMC operation status monitoring methods of the microcomputer host introduced above.

[0053] The embodiment of the present application also provides a computer program product. When the computer program product is run on a computer device, the computer device can execute any one of the BMC operation status monitoring methods for a microcomputer host described above.

[0054] The embodiment of the present application also provides a computer-readable storage medium, in which a computer program code is stored. When the computer program code is executed on a computer device, the computer device can execute any one of the BMC operation status monitoring methods for a microcomputer host described above.

[0055] In the embodiments provided in the present application, it should be understood that the provided computer device, computer program product and computer-readable storage medium are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the methods provided above and will not be repeated here.

[0056] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for monitoring the BMC operation status of a microcomputer host, characterized in that: The method comprises: During the operation of the microcomputer host, the prior alarm time, CPU temperature curve, radiator speed curve and CPU utilization curve of each monitoring time period monitored by the BMC are collected; the preset de-jitter time window and the CPU over-temperature start time when the CPU temperature exceeds the preset temperature threshold for the first time in each monitoring time period are obtained; Determine the degree of temperature anomaly based on the change correlation between the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve; determine the degree of heat dissipation anomaly based on the matching of the CPU temperature change during the radiator speed increase process; determine the alarm risk of each monitoring time period based on the overall stability of the CPU utilization curve, the temperature anomaly degree and the heat dissipation anomaly degree; In each monitoring time period, the alarm response requirement level of each monitoring time period is determined according to the time length between the prior alarm moment and the CPU overtemperature start moment, the CPU temperature distribution and the overall CPU temperature; Determine the window adjustment coefficient of each monitoring time period according to the alarm response requirement degree and the alarm risk; adjust the preset de-jitter time window according to the window adjustment coefficient to determine the dynamic de-jitter time window of each monitoring time period; monitor the BMC operation status of the microcomputer host according to the dynamic de-jitter time window.

2. The method for monitoring the BMC operation status of a microcomputer host according to claim 1, characterized in that: The process of obtaining the temperature anomaly degree includes: On the CPU temperature curve, the slope value of the line between the data point corresponding to each sampling moment and the data point corresponding to the previous sampling moment is used as the temperature slope value at each sampling moment; and the temperature rising trend characteristic value is determined according to the average value of the temperature slope values ​​at all sampling moments; On the CPU temperature curve, negative correlation mapping is performed between the mean value of the CPU temperature at all sampling moments and the difference between the preset temperature threshold value to determine the proximity degree of the threshold temperature; Determine the degree of correlation between the temperature and the rotation speed according to a negative correlation mapping of the Pearson correlation coefficient between the CPU temperature curve and the heat sink rotation speed curve; The degree of temperature anomaly in each monitoring time period is determined according to the temperature-speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value; the temperature-speed correlation degree is negatively correlated with the temperature anomaly degree; the threshold temperature proximity degree and the temperature rising trend characteristic value are both positively correlated with the temperature anomaly degree.

3. The method for monitoring the BMC operation status of a microcomputer host according to claim 2, characterized in that: The process of determining the degree of temperature anomaly in each monitoring time period according to the temperature-rotation speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value includes: The product of the negative correlation mapping value of the temperature-rotation speed correlation degree, the threshold temperature proximity degree and the temperature rising trend characteristic value is normalized to determine the temperature anomaly degree of each monitoring time period.

4. The method for monitoring the BMC operation status of a microcomputer host according to claim 2, characterized in that: The process of obtaining the abnormal degree of heat dissipation includes: On the radiator speed curve, the slope value of the line between the data point corresponding to each sampling moment and the data point corresponding to the previous sampling moment is used as the speed slope value at each sampling moment; the sampling moment when the speed slope value is greater than 0 is used as the speed rising moment; The difference between the speed slope value and the corresponding temperature slope value at each speed rising moment is negatively correlated to determine the instantaneous heat dissipation fault degree at each speed rising moment; the heat dissipation abnormality degree of each monitoring time period is determined based on the average of the instantaneous heat dissipation fault degrees at all speed rising moments.

5. The method for monitoring the BMC operation status of a microcomputer host according to claim 1, characterized in that: The process of obtaining the alarm risk includes: The CPU utilization standard deviations at all sampling moments on the CPU utilization curve are negatively correlated to determine the stability of the CPU utilization in each monitoring time period; The product of the negative correlation mapping value of the CPU utilization stability, the temperature abnormality and the heat dissipation abnormality is normalized to determine the alarm risk of each monitoring time period.

6. The method for monitoring the BMC operation status of a microcomputer host according to claim 1, characterized in that: The process of obtaining the alarm response requirement degree includes: The monitoring time period in which both the priori alarm moment and the CPU overtemperature start moment exist and the CPU overtemperature start moment is before the priori alarm moment is used as the analysis time period; the alarm response requirement degree of other monitoring time periods outside the analysis time period is set to the preset requirement degree; In each analysis time period, all sampling moments between the CPU over-temperature start moment and the prior alarm moment are taken as delay moments; the delay moment when the corresponding CPU temperature is greater than the preset temperature threshold is taken as the reference abnormal moment; the difference between the CPU temperature at each reference abnormal moment and the preset temperature threshold is taken as the instantaneous over-temperature degree at each reference abnormal moment; the overall over-temperature degree of each analysis time period is determined based on the accumulated value of the instantaneous over-temperature degrees at all reference abnormal moments; The product between the number of delay moments and the overall over-temperature degree in each analysis time period is normalized to determine the alarm response requirement degree in each analysis time period.

7. The method for monitoring the BMC operation status of a microcomputer host according to claim 1, characterized in that: The process of obtaining the window adjustment coefficient includes: A negative correlation mapping is performed on the product of the alarm response requirement degree and the alarm risk to determine the window adjustment coefficient of each monitoring time period.

8. The method for monitoring the BMC operation status of a microcomputer host according to claim 1, characterized in that: The process of acquiring the dynamic de-jittering time window includes: The dynamic de-jitter time window of each monitoring time period is determined by multiplying the product of the window adjustment coefficient and the preset de-jitter time window.

9. The method for monitoring the BMC operation status of a microcomputer host according to claim 1, characterized in that: The process of monitoring the BMC operating status of the microcomputer host according to the dynamic de-jittering time window includes: The dynamic de-jitter time window of each monitoring time period is used as the final de-jitter time window of each subsequent monitoring time period until a new dynamic de-jitter time window is re-determined when an alarm occurs; the BMC operating status of the microcomputer host is monitored based on the final de-jitter time window of each monitoring time period combined with the corresponding CPU temperature.

10. A BMC operation status monitoring system for a microcomputer host, characterized in that: The system comprises: The data collection preprocessing module is used to collect the prior alarm time, CPU temperature curve, radiator speed curve and CPU utilization curve of each monitoring time period monitored by the BMC during the operation of the microcomputer host; obtain the preset de-jitter time window and the CPU over-temperature start time when the CPU temperature exceeds the preset temperature threshold for the first time in each monitoring time period; The first determination module is used to determine the temperature abnormality according to the change correlation between the CPU temperature curve and the radiator speed curve and the overall rising trend of the CPU temperature in the CPU temperature curve; determine the heat dissipation abnormality according to the matching of the CPU temperature change during the radiator speed increase process; determine the alarm risk of each monitoring time period according to the overall stability of the CPU utilization curve, the temperature abnormality and the heat dissipation abnormality; A second determination module is used to determine the alarm response requirement degree of each monitoring time period according to the time length between the prior alarm moment and the temperature abnormality moment, the CPU temperature distribution and the overall CPU temperature in each monitoring time period; An operation status monitoring module is used to determine the window adjustment coefficient of each monitoring time period according to the alarm response requirement degree and the alarm risk; adjust the preset de-jitter time window according to the window adjustment coefficient to determine the dynamic de-jitter time window of each monitoring time period; and monitor the BMC operation status of the microcomputer host according to the dynamic de-jitter time window.

Images