Resource monitoring method, device and equipment

Abstract

The embodiment of the invention relates to the technical field of server resource management, in particular to a resource monitoring method, device and equipment. The resource monitoring method comprises the following steps: firstly, analyzing correlation among sensors according to measurement information of the sensors, performing clustering analysis on the sensors according to the correlation among the sensors, and grouping the sensors according to a clustering analysis result; meanwhile, according to the measurement information of the grouped sensors, a key sensor group and a non-key sensor group are found out, a relatively high monitoring frequency is set for the key sensor group, and a relatively low monitoring frequency is set for the non-key sensor group, so that the resource monitoring cost is reduced, and the resource monitoring pressure of the BMC is also relieved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of server resource management, and specifically to a resource monitoring method, apparatus and device. Background Art

[0002] With the rapid development of the Internet industry, the number of servers deployed worldwide has become very large, so the security of servers has gradually become a research focus. BMC (Baseboard Management Controller) chips are often deployed in mid-to-high-end servers. BMC chips have a series of monitoring and control functions, and the objects they monitor and control are system hardware. For example, they monitor the system's CPU, hard disk, network, voltage, fan, power supply, etc., and make corresponding adjustments to ensure that the system is in a healthy state.

[0003] As the server's steward, BMC holds a lot of system status information. As server functions become more and more powerful, more and more devices need to be monitored. The resources held by a single BMC chip are gradually unable to meet the needs of motherboard monitoring. Summary of the invention

[0004] In view of the above problems, the embodiments of the present application propose a resource monitoring method, device and equipment to solve the above technical problems existing in the prior art.

[0005] In one aspect of an embodiment of the present application, a resource monitoring method is provided, comprising:

[0006] Get the standard read and write duration information corresponding to multiple sensor resources;

[0007] Determine the number of groups M for grouping the plurality of sensor resources according to the standard read / write duration information, where M is a positive integer;

[0008] Acquire measurement information corresponding to a plurality of the sensor resources;

[0009] Determine the impact factors corresponding to the plurality of sensor resources according to the measurement information corresponding to the plurality of sensor resources;

[0010] Performing cluster analysis on the influencing factors corresponding to the plurality of sensor resources according to the grouping number M, dividing the plurality of sensor resources into M sensor resource groups, and determining the influencing factors corresponding to the plurality of sensor resource groups;

[0011] Determining monitoring periods corresponding to the plurality of sensor resource groups according to the influencing factors corresponding to the plurality of sensor resource groups;

[0012] The sensor resources in the multiple sensor resource groups are monitored respectively according to the monitoring periods corresponding to the multiple sensor resource groups.

[0013] In another aspect of the embodiment of the present application, a resource monitoring device is further provided, comprising:

[0014] Sensor resource acquisition module: used to obtain standard read and write duration information and measurement information corresponding to multiple sensor resources;

[0015] A grouping number determination module: used to determine the number of groups M for grouping the multiple sensor resources according to the standard read and write duration information corresponding to the multiple sensor resources, where M is a positive integer;

[0016] An impact factor determination module: configured to determine the impact factors corresponding to the plurality of sensor resources according to the measurement information corresponding to the plurality of sensor resources;

[0017] A sensor grouping module: configured to perform cluster analysis on the influencing factors corresponding to the plurality of sensor resources according to the grouping number M, divide the plurality of sensor resources into M sensor resource groups, and determine the influencing factors corresponding to the plurality of sensor resource groups;

[0018] A monitoring period determination module: used to determine the monitoring periods corresponding to the plurality of sensor resource groups according to the influencing factors corresponding to the plurality of sensor resource groups;

[0019] Resource monitoring module: used for monitoring the sensor resources in the plurality of sensor resource groups respectively according to the monitoring periods corresponding to the plurality of sensor resource groups.

[0020] In another aspect of the embodiment of the present application, a resource monitoring device is further provided, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0021] The memory is used to store executable instructions, and the executable instructions enable the processor to execute the operations of the resource monitoring method described in the above embodiment.

[0022] On the other hand, an embodiment of the present application further proposes a computer-readable storage medium, characterized in that at least one executable instruction is stored in the storage medium, and when the executable instruction is run on a resource monitoring device, the resource monitoring device executes the operation of the resource monitoring method described in any embodiment.

[0023] The resource monitoring method, apparatus and device proposed in the embodiments of the present application analyze the correlation between the sensors based on the measurement information of the sensors, perform cluster analysis on the sensors based on the correlation between the sensors, and group the sensors according to the results of the cluster analysis; at the same time, based on the measurement information of the grouped sensors, find out the key sensor group and the non-key sensor group, set a higher monitoring frequency for the key sensor group, and set a lower monitoring frequency for the non-key sensor group. In this way, the cost of resource monitoring is reduced, and the pressure on the BMC to perform resource monitoring is also alleviated.

[0024] The above description is only an overview of the technical solution of the embodiment of the present application. In order to more clearly understand the technical means of the embodiment of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present application. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings:

[0026] Figure 1 A schematic diagram of an application scenario of the resource monitoring method provided in an embodiment of the present application is shown;

[0027] Figure 2 A schematic diagram of the resource monitoring method provided in an embodiment of the present application is shown;

[0028] Figure 3 A schematic diagram of a sensor resource measurement information acquisition process provided in an embodiment of the present application is shown;

[0029] Figure 4 A schematic diagram of a sensor resource impact factor determination process provided by an embodiment of the present application is shown;

[0030] Figure 5 A schematic diagram of a resource monitoring device provided in an embodiment of the present application is shown;

[0031] Figure 6 A schematic diagram of the structure of a resource monitoring device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0033] In current servers, there are multiple hardware components, such as CPU, hard disk, network, power supply, motherboard, fan, etc. During the operation of the server, BMC needs to monitor various hardware resources through multiple sensors, such as the CPU operating temperature, fan speed, power supply power, motherboard voltage and current, etc. These sensors constitute sensor resources. The existing motherboard supports up to 8 CPUs. The heat dissipation of 8 CPUs requires 8 PWM controls, and the system fan also needs to be controlled. At the same time, the voltages of the CPU also need to be monitored. Each CPU needs to monitor the CPU input power supply voltage, CPU output power supply voltage, CPU core voltage, CPU auxiliary voltage, etc.

[0034] At present, the mainstream BMC sensor resources include 16 fan output PWMs, 16 fan monitoring Tachs, and 16 ADC voltage monitoring channels. These sensor resources are obviously not enough for high-performance servers. Therefore, it is necessary to expand sensor resources. Expanding sensor resources can be done by introducing new MCUs to monitor various resources through the newly added MCUs. However, such a design will first increase hardware costs. Secondly, after forming a master (BMC) slave (MCU) system, the communication stability between the master and slave systems will also be problematic. When a slave device fails, it will affect the data and operation of the master system. Similarly, the master system will also affect the slave system.

[0035] In view of this, the embodiment of the present application proposes a resource monitoring method, device and equipment to solve the above problems existing in the prior art. The resource monitoring scheme proposed in the embodiment of the present application first analyzes the correlation between the sensors according to the measurement information of each sensor, performs cluster analysis on these sensors according to the correlation between the sensors, and groups the sensors according to the results of the cluster analysis. At the same time, according to the measurement information of the grouped sensors, the key sensor group and the non-key sensor group are found, and a higher monitoring frequency is set for the key sensor group, and a lower monitoring frequency is set for the non-key sensor group. When there are too many ADCs, PWMs, and TACHs in the BMC system, and there are not enough channels in the BMC chip for data reading, the resource monitoring method provided by the embodiment of the present application is used to group the sensors, and the polling mechanism of sensor monitoring is used to read the sensors in groups. When a certain group of sensors needs to be read, the corresponding monitoring channel is switched through the corresponding GPIO interface, and then the data can be read. At the same time, the key sensor group and the non-key sensor group are found, and the key sensor group is monitored at a high frequency according to the determined monitoring cycle, and the monitoring frequency is reduced for the non-key sensor group. In this way, the cost of resource monitoring is reduced and the pressure of BMC for resource monitoring is reduced.

[0036] like Figure 1As shown, a schematic diagram of the application scenario of the resource monitoring method proposed in the embodiment of the present application is shown, wherein BMC-Chip is the BMC system main control chip, which is a key component in the server hardware for monitoring and managing the server. It is usually integrated on the server motherboard, has its own processor, memory and storage, can run independently of the server main processor, and realizes the monitoring, management and maintenance of the server hardware and system. BMC-Chip communicates and controls with the server through the bus, and has multiple functions, including but not limited to device information management, server status monitoring, remote control management and maintenance management.

[0037] Data lines AE are used to input data. These data lines are channels for BMC to monitor data readings. They are connected to different sensor resources, such as hard disks, network processors, CPUs, etc. BMC-Chip obtains relevant information of corresponding hardware devices through corresponding data buses. In the embodiment of the present application, data line F is used to transmit filtered and processed data. When receiving an instruction, the switching chip will select a group of channels according to the instruction. This group of data lines can be regarded as directly connected to data line F.

[0038] The switching chip is used to select one channel of the data lines AE according to the input of the control line, and connect this channel to the data line F, which can be regarded as directly physically connecting each sensor resource to the BMC-Chip.

[0039] The control lines are output by the GPIO of the BMC-Chip, and the channel switching of the switching chip is controlled by these GPIO control lines.

[0040] The resource monitoring method proposed in this application is usually run through the BMC-Chip, sending operation instructions to the switching chip to monitor the selected sensor resources according to a specific monitoring cycle, without the need to add new processors, saving costs. At the same time, as the number of resources to be monitored increases, the load will not increase, greatly improving the monitoring efficiency of sensor resources.

[0041] In order to clearly describe the resource monitoring method proposed in the embodiment of the present application, Figure 2 The flowchart of the resource monitoring method proposed in the embodiment of the present application is shown, which specifically includes:

[0042] S100: Obtain standard read and write duration information corresponding to multiple sensor resources;

[0043] In the embodiment of the present application, sensor resources mainly refer to sensors used by users in the server to monitor various hardware and network resources. They are mainly used to monitor the CPU operating temperature, fan speed, power supply power, motherboard voltage and current, etc. They can all be collectively referred to as sensor resources.

[0044] Each sensor resource has standard read and write duration information when it leaves the factory. This standard read and write duration information is used to indicate the data reading speed of different types of sensor resources. For example, for I2C sensors, the standard read and write duration information is 170us; for GPIO sensor reading and writing, the standard read and write duration information is 10ns. In this step, it is necessary to obtain the standard read and write duration information corresponding to multiple sensor resources that need to be monitored in advance. For the same type of sensor resources, they generally have the same standard read and write duration information.

[0045] S200: Determine the number of groups M for grouping the plurality of sensor resources according to the standard read / write duration information, where M is a positive integer;

[0046] After obtaining the standard read and write duration information of all sensor resources to be monitored, the total read and write duration information of all sensor resources to be monitored is determined.

[0047] Assume that each type of sensor resource has Ki sensors, i represents different types of sensor resources, and the standard read and write duration information of the sensor resource of this type is ti, then the read and write duration Ti of all sensor resources of this type is:

[0048] Ti = Ki*ti;

[0049] Assuming there are m groups of sensor resources, the total read and write duration information T of all sensor resources to be monitored is:

[0050] T=T0+T1+T2+……+Tm;

[0051] In order to group multiple sensor resources, it is necessary to first determine the number of groups. In the embodiment of the present application, the total read and write time information T of all sensor resources to be monitored and the maximum standard read and write time information T of all sensor resources to be monitored can be used to determine the number of groups. max To determine the number of groups.

[0052] Assume that among all sensor resources to be monitored, the maximum standard read and write time of the sensor is T max , the minimum standard read and write time information is T min , then determine the number of groups M as:

[0053] M=T / T max ;

[0054] According to the above formula, by dividing T and T max After the quotient between them is rounded up, the number M of groups for grouping the multiple sensor resources can be determined, and M is a positive integer.

[0055] S300: Acquire measurement information corresponding to a plurality of sensor resources;

[0056] Before monitoring sensor resources, first obtain measurement information of multiple sensor resources under different system states, where the system states include different ambient temperatures, different pressures, etc. For example, when the ambient temperature is A, pressurize different server hardware structures respectively, and obtain measurement information of each sensor resource under different pressurization conditions. Each set of data corresponds to different measurement information, which serves as the data basis for BMC-Chip to perform resource monitoring and analysis.

[0057] In order to accurately group sensor resources and find out the main influencing factors of system fluctuations, in an embodiment of the present application, sensor data in various states are tested, collected and sorted, and data needs to be collected in multiple environments.

[0058] Before testing, collecting and collating the sensor data, it is necessary to ensure that the sensor resources are in a stable state. In the embodiment of the present application, before testing, collecting and collating the sensor resource data, the minimum standard read and write time information T is determined from the standard read and write time information corresponding to the plurality of sensor resources. min and the total read and write time information T, the total read and write time information T and the minimum standard read and write time information T min The quotient is rounded to obtain the number of steady-state times L. If the change in the measurement information of the plurality of sensor resources is less than or equal to the preset threshold value within the L*T time length, it means that each of the sensor resources is in a steady-state state, and then the test data collection for each sensor resource is started. If the change in the measurement information of the plurality of sensor resources is greater than the preset threshold value, it means that the plurality of sensor resources are still in a non-steady-state state, and then the waiting continues until the plurality of sensor resources are in a steady-state state.

[0059] When the sensor resources are in a steady state, the multiple sensor resources are sampled and taken as sample values ​​for analysis. For the convenience of explanation, in the embodiment of the present application, the system environment is constructed with the processor, hard disk and network as the main hardware resources, and the measurement information of the sensor resources is obtained under different pressure conditions as an example for explanation.

[0060] Specifically, the method for obtaining the measurement information corresponding to the plurality of sensor resources in step S300 is as follows: Figure 3 As shown, including:

[0061] S301: Obtaining upper and lower operating temperature values ​​of the processor, network, and hard disk;

[0062] In actual work, various hardware devices have standard operating temperature ranges when they leave the factory, including the upper and lower operating temperature values. Taking sensors, networks, and hard disks as examples, they include the upper and lower operating temperature values ​​of the processor, the upper and lower operating temperature values ​​of the network, and the upper and lower operating temperature values ​​of the hard disk. By placing the above hardware devices in different working environments, different sensor measurement information can be obtained.

[0063] S302: Configure the working environment temperature to be: working temperature lower limit + preset temperature step * X;

[0064] In the embodiment of the present application, data sampling is performed on the measurement information of the sensor resources at multiple working environment temperatures. Starting from the lower limit of the working temperature determined in the above steps, the working environment temperature is gradually increased according to the preset temperature step, and testing and data acquisition are performed at each working environment temperature.

[0065] Assume that, in the embodiment of the present application, the preset temperature step is 1°C, and the lower limit of the working temperature is 30°C. Then, the working environment temperature is first set to 30°C, and S303-S312 is executed; each time it is executed, the working environment temperature is increased by 1°C, and X is the number of cycles.

[0066] S303: making the processor idle, and the network and the hard disk have no tasks; respectively obtaining measurement information of the plurality of sensors;

[0067] In actual testing, first ensure that the processor is unloaded, the network and the hard disk have no tasks, and when the system is in a stable state, in this scenario, obtain the measurement information of the multiple sensors respectively. The determination of the stable state can be carried out in the manner described in the above embodiment, that is, if the change in the measurement information of the multiple sensor resources is less than or equal to the preset threshold within the L*T time length, it means that each of the sensor resources is in a stable state.

[0068] S304: gradually pressurizing the processor with a preset processor pressurization step length; respectively obtaining measurement information of the plurality of sensors;

[0069] After obtaining the measurement information of multiple sensors when the processor is unloaded and the network and the hard disk are in a non-task state, the processor is first pressurized according to a preset processor pressurization step length, and the measurement information of multiple sensors under corresponding pressures is obtained respectively.

[0070] The preset processor pressure step length may be 1%, and the processor is pressurized step by step in 1% steps;

[0071] The processor is pressurized to 1%, and when the system is in a stable state, measurement information of the plurality of sensors in the 1% state is obtained by polling;

[0072] Sequentially, the processor is pressurized to 2%, and when the system is in a stable state, measurement information of the plurality of sensors in the 2% state is obtained by polling;

[0073] The processor is pressurized to 3%, and when the system is in a stable state, measurement information of the plurality of sensors in the 3% state is obtained by polling;

[0074] By analogy, the processor can be pressurized to its extreme state, such as 100% state, and the measurement information of the plurality of sensors under the extreme pressure state can be obtained.

[0075] S305: gradually decompressing the processor with a preset processor decompression step; respectively obtaining measurement information of the plurality of sensors;

[0076] When the processor is pressurized to its limit state, the processor is gradually depressurized according to the preset processor decompression step length, and measurement information of multiple sensors is obtained respectively, and the preset processor decompression step length is 1% as an example for explanation;

[0077] The processor is decompressed to 99%, and when the system is in a stable state, the measurement information of each of the sensors in the 99% state is obtained by polling;

[0078] Sequentially, the processor is increased or decreased to 98%, and when the system is in a stable state, the measurement information of the plurality of sensors in the 98% state is obtained by polling;

[0079] Decompressing the processor to 97%, and when the system is in a stable state, polling to obtain measurement information of the plurality of sensors in the 97% state;

[0080] By analogy, the processor can be decompressed to a no-load state, and the measurement information of the plurality of sensors in the no-load state can be obtained. The purpose here is to compare the state changes of the system after different operations.

[0081] S306: making the processor idle, and the network and the hard disk have no tasks; respectively obtaining measurement information of the plurality of sensors;

[0082] After the gradual stress test on the processor is completed, the system is put into a stable state, and the measurement information of each sensor resource is obtained. This step is the same as S303 and will not be described in detail here.

[0083] S307: gradually pressurizing the network with a preset network pressurization step; respectively obtaining measurement information of a plurality of sensors;

[0084] After obtaining the measurement information of each sensor resource when the processor is unloaded and the network and the hard disk are in a non-task state, the network begins to be pressurized according to a preset network pressurization step, and the measurement information of multiple sensors under corresponding pressures is obtained respectively.

[0085] The preset network pressure step size may be 1% of the upper limit of the network bandwidth, and the network is gradually pressured in steps of 1%;

[0086] The network bandwidth is increased to 1%, and when the system is in a stable state, measurement information of the plurality of sensors in the 1% state is obtained by polling;

[0087] Sequentially, the network bandwidth is compressed to 2%, and when the system is in a stable state, measurement information of the plurality of sensors in the 2% state is obtained by polling;

[0088] The network bandwidth is compressed to 3%, and when the system is in a stable state, measurement information of the plurality of sensors in the 3% state is obtained by polling;

[0089] By analogy, the network bandwidth can be pressurized to its limit state, such as 100%, and the measurement information of the multiple sensors under the limit pressure state can be obtained.

[0090] S308: gradually reducing the network pressure with a preset network pressure reduction step; respectively obtaining measurement information of a plurality of sensors;

[0091] When the network bandwidth is pressurized to its limit state, the network bandwidth is gradually depressurized according to the preset network decompression step, and the measurement information of the multiple sensors is obtained respectively, and the preset network decompression step is 1% of the upper limit of the network bandwidth as an example for explanation;

[0092] The network bandwidth is reduced to 99%, and when the system is in a stable state, measurement information of the plurality of sensors in the 99% state is obtained by polling;

[0093] Sequentially, the network bandwidth is decompressed to 98%, and when the system is in a stable state, measurement information of the plurality of sensors in the 98% state is obtained by polling;

[0094] The network bandwidth is reduced to 97%, and when the system is in a stable state, measurement information of the plurality of sensors in the 97% state is obtained by polling;

[0095] By analogy, the network bandwidth can be reduced to a no-task state, and the measurement information of the plurality of sensors in the no-task state can be obtained.

[0096] S309: making the processor idle, and the network and the hard disk have no tasks; respectively obtaining measurement information of the plurality of sensors;

[0097] After the gradual pressure test on the network bandwidth is completed, the system is put into a stable state, and the measurement information of each sensor resource is obtained. This step is the same as S303 and will not be repeated here.

[0098] S310: gradually pressurizing the hard disk with a preset hard disk pressurizing step length; respectively obtaining measurement information of a plurality of sensors;

[0099] After obtaining the measurement information of multiple sensors when the processor is unloaded and the network and the hard disk are in a non-task state, start pressurizing the hard disk according to the preset hard disk pressurization step, and obtain the measurement information of multiple sensors under corresponding pressures.

[0100] The preset hard disk pressure step length may be 1%, and the hard disk is gradually pressurized in steps of 1% of the upper limit of the hard disk transmission speed;

[0101] Pressurize the hard disk to 1%, and when the system is in a stable state, poll to obtain measurement information of the multiple sensors in the 1% state;

[0102] Sequentially, the hard disk is pressurized to 2%, and when the system is in a stable state, measurement information of the plurality of sensors in the 2% state is obtained by polling;

[0103] Pressurizing the hard disk to 3%, and when the system is in a stable state, polling to obtain measurement information of the plurality of sensors in the 3% state;

[0104] By analogy, the hard disk can be pressurized to its extreme state, such as 100% state, and the measurement information of the multiple sensors under the extreme pressure state can be obtained.

[0105] S311: gradually decompressing the hard disk with a preset hard disk decompression step length; respectively obtaining measurement information of a plurality of sensors;

[0106] When the hard disk is pressurized to its limit state, the hard disk is gradually depressurized according to the preset hard disk decompression step length, and the measurement information of the multiple sensors is obtained respectively, and the preset hard disk decompression step length is taken as 1% of the upper limit of the hard disk transmission speed as an example for explanation;

[0107] Decompressing the hard disk to 99%, and when the system is in a stable state, polling to obtain measurement information of the plurality of sensors in the 99% state;

[0108] Sequentially, the hard disk is decompressed to 98%, and when the system is in a stable state, measurement information of the plurality of sensors in the 98% state is obtained by polling;

[0109] Decompressing the hard disk to 97%, and when the system is in a stable state, polling to obtain measurement information of the plurality of sensors in the 97% state;

[0110] By analogy, the hard disk can be decompressed to a non-task state, and measurement information of the multiple sensors in the non-task state can be obtained.

[0111] S312: Determine whether the working environment temperature is less than the working temperature upper limit;

[0112] After completing the operations of steps S303-S311, it is determined whether the working environment temperature is less than the working temperature upper limit. If so, the process proceeds to step S302 to continue testing each sensor resource and obtain measurement information under each test condition. If the working environment temperature is greater than or equal to the working temperature upper limit, the test of the sensor resource is completed and the process ends.

[0113] According to the description in the above steps, assuming that the system includes two sensor resources X, Y, and Z, and records data five times, the format of the recorded data is shown in Table 1:

[0114] Table 1

[0115]

[0116]

[0117] Among them, X, Y, and Z represent different sensors, and S1-S5 are data records obtained at different time points and under different pressure conditions.

[0118] By acquiring the measurement information of each sensor resource through the above method proposed in the embodiment of the present application, the measurement information of the sensor resources under various conditions can be polled as much as possible, the sensor resources can be grouped more accurately, and the efficiency of resource monitoring can be improved.

[0119] After the measurement information corresponding to the plurality of sensor resources is acquired through the above step S300 and steps S301 - S312 , step S400 is continued to be executed.

[0120] S400: Determine the impact factors corresponding to the plurality of sensor resources according to the measurement information corresponding to the plurality of sensor resources;

[0121] In order to group the multiple sensor resources, the embodiment of the present application performs correlation analysis on the measurement information of the multiple sensor resources to determine the influencing factors between the multiple sensor resources. The influencing factors are used to describe the complex relationship between multiple variables and represent the correlation between the multiple sensor resources.

[0122] In the embodiment of the present application, the measurement information corresponding to the plurality of sensor resources is analyzed by a factor analysis method to determine the impact factors corresponding to the plurality of sensor resources. Specifically, the method for determining the impact factors corresponding to the plurality of sensor resources according to the measurement information corresponding to the plurality of sensor resources in step S400 is as follows: Figure 4 As shown:

[0123] S401: Determine an average value of measurement information corresponding to the plurality of sensor resources according to measurement information of the plurality of sensor resources;

[0124] Assuming that the number of samples in each dimension is N, the mean of each dimension is obtained by the following formula:

[0125]

[0126] Take the data in Table 1 as an example, where each dimension corresponds to a sensor resource. In Table 1, sensor resources X and Y are included. Therefore, it includes two dimensions, and its mean μ is:

[0127] μ x =7

[0128] μ y =6

[0129] μ z =8

[0130] S402: determining the variance and standard deviation of the measurement information corresponding to the plurality of sensor resources according to the average value of the measurement information corresponding to the plurality of sensor resources;

[0131] variance It is a measure of the degree of dispersion of a single random variable or a set of data. The mathematical expression is as follows:

[0132]

[0133] n represents the sample size, Represents the mean.

[0134] Standard deviation x is the arithmetic square root of the variance:

[0135]

[0136] Taking the sample values ​​in Table 1 as an example, the calculated standard deviation is:

[0137]

[0138] S403: performing standardization processing on the variance and standard deviation of the measurement information corresponding to the plurality of sensor resources to obtain standardized measurement information corresponding to the plurality of sensor resources;

[0139] When variables have different dimensions and numerical ranges, standardization helps eliminate these differences, making the results of factor analysis more fair and reliable. At the same time, standardized data can improve the stability of factor analysis, especially when the numerical ranges of variables vary greatly.

[0140] Standardize the sample data:

[0141]

[0142] Taking the sample values ​​in Table 1 as an example, the standardized data is:

[0143] X1: (8-7) / 1.58 X2: (6-7) / 1.58 X3: (7-7) / 1.58 X4: (9-7) / 1.58 X5: (5-7) / 1.58

[0144] Y1: (7-6) / 1.58 Y2: (5-6) / 1.58 Y3: (6-6) / 1.58 Y4: (8-6) / 1.58 Y5: (4-6) / 1.58

[0145] Z1: (9-8) / 1.58 Z2: (8-8) / 1.58 Z3: (7-8) / 1.58 Z4: (10-8) / 1.58 Z5: (6-8) / 1.58 The standardized measurement information is shown in Table 2:

[0146] Table 2

[0147] X Y Z S1 0.63 0.63 0.63 S2 -0.63 -0.63 0 S3 0 0 -0.63 S4 1.27 1.27 1.27 S5 -1.27 -1.27 -1.27

[0148] S404: determining a correlation coefficient between the plurality of sensor resources according to standardized measurement information corresponding to the plurality of sensor resources;

[0149] r xy Represents the correlation coefficient in the XY dimension:

[0150]

[0151] Where k represents the kth group of data, x k is the kth group of data on the X dimension after standardization;

[0152] Substituting the standardized data in Table 2 into the above formula, the correlation coefficient r is obtained. Since the effects of using standardized data and original data when calculating the correlation coefficient matrix are mathematically consistent, for the convenience of calculation, the original data is used for calculation in this example:

[0153]

[0154] Calculate the correlation coefficient: r xy =0.9r xz =0.9r yz =0.9.

[0155] S405: generating a correlation coefficient matrix according to the correlation coefficients between the plurality of sensor resources;

[0156] The above correlation coefficient r generates a correlation coefficient matrix:

[0157]

[0158] Right now:

[0159]

[0160] S406: generating a plurality of feature vectors corresponding to the sensor resources according to the correlation coefficient matrix;

[0161] Calculate the above correlation coefficient matrix and take the eigenvalue λ i and the eigenvector V i , i represents the i-th group, and the size of the eigenvalue λ indicates the size of the overall system variance it explains.

[0162] From the above correlation coefficient matrix, we can estimate:

[0163] λ1=2.8,V1=[1 1 1]

[0164] λ2=0.1,V2=[-1 1 0]

[0165] λ3=0.1,V3=[-1 0 1]

[0166] From the above, we can see that λ1 produces a variance of 2.8 units; λ2 produces a variance of 0.1 units; λ3 produces a variance of 0.1 units;

[0167] Its relationship:

[0168] λ1=X+Y+Z

[0169] λ2=-1X+Y

[0170] λ3=-1X+Z

[0171] S407: Generate a load matrix according to the characteristic vectors corresponding to the plurality of sensor resources;

[0172] Generate a load matrix using V1, V2, and V3:

[0173]

[0174] S408: Generate impact factors corresponding to a plurality of the sensor resources according to the load matrix.

[0175] In order to more clearly illustrate how to generate the impact factor through the load matrix, the present embodiment uses the following data as an example:

[0176] Assume that the acquired data is sorted and analyzed according to the above steps to obtain a set of factors and correlation coefficients, and obtain the corresponding eigenvalues:

[0177] λ1=2.8,λ2=0.1,λ3=0.1

[0178] The corresponding eigenvectors are obtained:

[0179] λ1=X+Y+Z

[0180] λ2=-1X+Y

[0181] λ3=-1X+Z

[0182] By factoring it, we get the coefficient equation system:

[0183] S1=λ1X+λ1Y+λ1Z

[0184] S2=-λ2X+λ2Y

[0185] S3=-λ3X+λ3Z

[0186] Substituting λ1=2.8,λ2=0.1,λ3=0.1, we get:

[0187] S1=2.8X+2.8Y+2.8Z

[0188] S2=-0.1X+0.1Y

[0189] S3=-0.1X+0.1Z

[0190] All factors are combined to obtain an impact factor that can reflect the overall impact, where S is the system impact element, which is used to amplify the impact on the system and facilitate subsequent clustering operations:

[0191] S=2.6X+2.9Y+2.9Z

[0192] It can be obtained that the impact factor corresponding to sensor resource X is 2.6, the impact factor corresponding to sensor resource Y is 2.9, and the impact factor corresponding to sensor resource Z is 2.9.

[0193] Through the above step S400 and steps 401-S408, the impact factors corresponding to the multiple sensor resources are determined. These impact factors reflect the correlation between different sensor resources. Through the impact factors corresponding to these sensor resources, multiple sensor resources can be grouped and monitored, thereby improving the efficiency of resource monitoring.

[0194] Furthermore, since the above-mentioned load matrix is ​​not unique, in the embodiment of the present application, it is also proposed to rotate the load matrix, and generate multiple influencing factors corresponding to the sensor resources according to the rotated load matrix. The purpose of rotating the load matrix is ​​to make each variable have a relatively high load on as few factors as possible, so that the load of a certain variable on a certain factor tends to 1, and the load on other factors tends to 0, even if the square value of the elements of each column or row of the load matrix is ​​polarized to 0 and 1.

[0195] In the embodiment of the present application, the loading matrix can be selected by Varimax method. Varimax rotation is a commonly used orthogonal rotation method, the purpose of which is to simplify the loading matrix so that each factor has a high loading with only a few variables and a low loading with other variables. The dimension of the rotated loading matrix is ​​the same as the original loading matrix. Rotation does not change the basic relationship between factors and dimensions, but adjusts the specific loading values.

[0196] After the impact factors corresponding to the plurality of sensor resources are determined through the above steps S400 and steps 401 to S408, step S500 is continued to be executed.

[0197] S500: performing cluster analysis on the influencing factors corresponding to the plurality of sensor resources according to the grouping number M, dividing the plurality of sensor resources into M sensor resource groups, and determining the influencing factors corresponding to the plurality of sensor resource groups;

[0198] Cluster analysis refers to calculating the distance between samples, merging the closest points into the same class each time. Then, the distance between classes is calculated, and the classes with the closest distance are merged into a large class. The merging is continued until a class is synthesized. The methods for calculating the distance between classes include: the shortest distance method, the longest distance method, the middle distance method, the class average method, etc. For example, the shortest distance method defines the distance between classes as the shortest distance between samples between classes. The strategy of cluster analysis is to first treat each object as a cluster, and then merge these atomic clusters into larger and larger clusters until all objects are in one cluster or a certain termination condition is met.

[0199] In the embodiment of the present application, the influencing factors corresponding to multiple sensor resources are clustered using the minimum distance to divide them into M sensor resource groups. By treating each sensor resource as a class and calculating the minimum distance between each of them, the Euclidean distance method can be used to merge the two classes with the smallest distance into a new class, and the mean of the new class is the value of the new class. The distance between the new class and all classes is recalculated until the number of class mergers M required is reached.

[0200] In order to more clearly illustrate the process of cluster analysis, the present application embodiment continues to illustrate with the above example. In the above embodiment, the impact factor is obtained according to S=2.6X+2.9Y+2.9Z:

[0201] [X, Y, Z]: [2.6, 2.9, 2.9]

[0202] Assume that the number of groups M is 2, that is, the above X, Y and Z sensor resources need to be divided into two groups, and then perform clustering operation on each group. Since the distance between Y and Z is the same, we get:

[0203] [X][Y, Z]→[2.6][2.9]

[0204] It can be obtained that the grouping method is that the Y and Z sensor resources form a sensor resource group, and the corresponding impact factor is 2.9; X forms a sensor resource group, and the corresponding impact factor is 2.6.

[0205] S600: Determine monitoring periods corresponding to the plurality of sensor resource groups according to influencing factors corresponding to the plurality of sensor resource groups;

[0206] After determining the grouping of sensor resources, the embodiment of the present application further proposes to determine different monitoring periods for each sensor resource group, and to monitor different sensor resource groups using different monitoring frequencies, thereby further reducing the monitoring load of the BMC-Chip.

[0207] The embodiment of the present application proposes to take the reciprocals of the impact factors corresponding to the multiple sensor resource groups, obtain the minimum reciprocal value from the multiple reciprocals, and use the minimum reciprocal value as the unit monitoring period; at the same time, the quotient values ​​of the reciprocals corresponding to the multiple sensor resource groups and the unit monitoring period are rounded off to determine the monitoring period corresponding to the multiple sensor resource groups.

[0208] Specifically, the above data is used as an example for explanation. In the above example, the impact factor of the determined sensor resource group is: [2.6][2.9]

[0210] Take the inverse of the impact factor corresponding to each sensor resource group:

[0211] 1 / 2.6=0.38;

[0212] 1 / 2.9=0.34;

[0213] It can be seen that the inverse of the impact factor of the sensor resource group composed of the Y and Z sensors is the minimum inverse value, that is, the unit monitoring period is determined to be 0.34.

[0214] Divide the inverse of the impact factor of the sensor resource group of sensor X, 0.38, by the unit monitoring period, 0.34:

[0215] 0.38 / 0.34=1.1;

[0216] After rounding up, the value is 2, which means that the monitoring period corresponding to the X sensor is determined to be 2.

[0217] Therefore, through the above method, it can be confirmed that the monitoring period corresponding to [X] is 2, and the monitoring period corresponding to [Y, Z] is 1, that is, the monitoring frequency of [X] is less than the monitoring frequency of [Y, Z].

[0218] S700: Monitor sensor resources in a plurality of sensor resource groups respectively according to monitoring periods corresponding to the plurality of sensor resource groups.

[0219] According to the above determined sensor grouping and the corresponding monitoring period, the sensor resources in the multiple sensor resources are monitored respectively, and the above example is continued as an example:

[0220] In the above steps, the sensors are grouped as [X][Y, Z], that is, the Y and Z sensors are one sensor group, and the corresponding monitoring cycle is 1; the X sensors are one group, and the corresponding monitoring cycle is 2. Then the BMC-Chip monitors the Y and Z sensors in every monitoring cycle, and monitors the X sensor every 2 monitoring cycles, that is, the monitoring frequency of the Y and Z sensors is higher than the monitoring frequency of the X sensor.

[0221] It can be seen from the above embodiments that the resource monitoring method proposed in the embodiments of the present application first analyzes the correlation between the sensors based on the measurements of the sensors, clusters the sensors based on the correlation between the sensors, and groups the sensors based on the results of the cluster analysis; at the same time, based on the measurement information of the grouped sensors, find out the key sensor group and the non-key sensor group, set a higher monitoring frequency for the key sensor group, and set a lower monitoring frequency for the non-key sensor group. In this way, the cost of resource monitoring is reduced, and the pressure on the BMC to perform resource monitoring is also reduced.

[0222] Furthermore, the present application embodiment provides a resource monitoring device, such as Figure 5 As shown, the resource monitoring device 800 includes: a sensor resource acquisition module 801, a grouping quantity determination module 802, an impact factor determination module 803, a sensor grouping module 804, a monitoring period determination module 805 and a resource monitoring module 806;

[0223] The sensor resource acquisition module 801 is used to obtain standard read and write duration information and measurement information corresponding to multiple sensor resources;

[0224] The grouping number determination module 802 is used to determine the number of groups M for grouping the multiple sensor resources according to the standard read and write duration information corresponding to the multiple sensor resources, where M is a positive integer;

[0225] The impact factor determination module 803 is used to determine the impact factors corresponding to the multiple sensor resources according to the measurement information corresponding to the multiple sensor resources;

[0226] The sensor grouping module 804 is used to perform cluster analysis on the influencing factors corresponding to the multiple sensor resources according to the grouping number M, divide the multiple sensor resources into M sensor resource groups, and determine the influencing factors corresponding to the multiple sensor resource groups;

[0227] The monitoring period determination module 805 is used to determine the monitoring periods corresponding to the plurality of sensor resource groups according to the impact factors corresponding to the plurality of sensor resource groups;

[0228] The resource monitoring module 806 is used to monitor the sensor resources in the multiple sensor resource groups respectively according to the monitoring periods corresponding to the multiple sensor resource groups.

[0229] The grouping quantity determination module 802 is further used to determine the total read / write duration information T according to the standard read / write duration information corresponding to the multiple sensor resources; determine the maximum standard read / write duration information T from the standard read / write duration information corresponding to the multiple sensor resources. max ; The total read and write time information T and the maximum standard read and write time information T max After rounding the quotient between , the number of groups M for grouping the plurality of sensor resources is determined.

[0230] Furthermore, the sensor resource acquisition module 801 is further configured to determine the minimum standard read / write duration information T from the standard read / write duration information corresponding to the plurality of sensor resources. min ; The total read and write time information T and the minimum standard read and write time information T minAfter rounding the quotient, the steady-state number L is determined; if the change in the measurement information of the multiple sensor resources is less than the preset threshold within the L*T time length, the step of obtaining the measurement information corresponding to the multiple sensor resources is executed.

[0231] Furthermore, the sensor resource acquisition module 801 is also used to obtain the upper and lower operating temperature limits of the processor, network and hard disk; configure the working environment temperature in sequence according to the lower operating temperature limit and the preset temperature step, and perform the following steps at multiple working environment temperatures until the working environment temperature is greater than or equal to the upper operating temperature limit: make the processor unloaded, and the network and the hard disk have no tasks; obtain measurement information of multiple sensors respectively; gradually pressurize the processor with a preset processor pressurization step; obtain measurement information of multiple sensors respectively; gradually depressurize the processor with a preset processor decompression step; obtain multiple sensors respectively The method comprises the steps of: obtaining measurement information of a plurality of sensors; idling the processor, and making the network and the hard disk have no tasks; respectively obtaining measurement information of a plurality of sensors; gradually pressurizing the network with a preset network pressurization step length; respectively obtaining measurement information of a plurality of sensors; gradually depressurizing the network with a preset network decompression step length; respectively obtaining measurement information of a plurality of sensors; idling the processor, and making the network and the hard disk have no tasks; respectively obtaining measurement information of a plurality of sensors; gradually pressurizing the hard disk with a preset hard disk pressurization step length; respectively obtaining measurement information of a plurality of sensors; gradually depressurizing the hard disk with a preset hard disk decompression step length; respectively obtaining measurement information of a plurality of sensors.

[0232] Furthermore, the influencing factor determination module 803 is also used to determine the average value of the measurement information corresponding to the multiple sensor resources based on the measurement information of the multiple sensor resources; determine the variance and standard deviation of the measurement information corresponding to the multiple sensor resources based on the average value of the measurement information corresponding to the multiple sensor resources; standardize the variance and standard deviation of the measurement information corresponding to the multiple sensor resources to obtain standardized measurement information corresponding to the multiple sensor resources; determine the correlation coefficient between the multiple sensor resources based on the standardized measurement information corresponding to the multiple sensor resources; generate a correlation coefficient matrix based on the correlation coefficient between the multiple sensor resources; generate characteristic vectors corresponding to the multiple sensor resources based on the correlation coefficient matrix; generate a load matrix based on the characteristic vectors corresponding to the multiple sensor resources; generate influencing factors corresponding to the multiple sensor resources based on the load matrix.

[0233] Furthermore, the impact factor determination module 803 is further used to rotate the load matrix; and generate impact factors corresponding to a plurality of sensor resources according to the rotated load matrix.

[0234] Furthermore, the monitoring frequency determination module 804 is also used to take the inverses of the impact factors corresponding to the multiple sensor resource groups, obtain the minimum inverse value from the multiple inverses, and use the minimum inverse value as the unit monitoring period; after rounding the quotient of the inverses corresponding to the multiple sensor resource groups and the unit monitoring period, the monitoring period corresponding to the multiple sensor resource groups is determined.

[0235] It can be seen from the above embodiments that the resource monitoring device proposed in the embodiments of the present application first analyzes the correlation between the sensors based on the measurement information of the sensors, clusters the sensors based on the correlation between the sensors, and groups the sensors according to the results of the cluster analysis; at the same time, based on the measurement information of the grouped sensors, finds out the key sensor group and the non-key sensor group, sets a higher monitoring frequency for the key sensor group, and sets a lower monitoring frequency for the non-key sensor group. In this way, the cost of resource monitoring is reduced, and the pressure on the BMC to perform resource monitoring is also reduced.

[0236] Figure 6 A schematic diagram of the structure of a resource monitoring device proposed in an embodiment of the present application is shown. The embodiment of the present application does not limit the specific implementation of the resource monitoring device.

[0237] like Figure 6 As shown, the resource monitoring device may include: a processor (processor) 902 , a communication interface (Communications Interface) 904 , a memory (memory) 906 , and a communication bus 908 .

[0238] The processor 902, the communication interface 904, and the memory 906 communicate with each other via a communication bus 908. The communication interface 904 is used to communicate with other devices such as a client or other server network elements. The processor 902 is used to execute a program 910, which can specifically execute the relevant steps in the above-mentioned data transaction authority control method embodiment.

[0239] Specifically, the program 910 may include program codes, which include computer operation instructions.

[0240] The processor 902 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the server may be processors of the same type, such as one or more CPUs, or processors of different types, such as one or more CPUs and one or more ASICs.

[0241] The memory 906 is used to store the program 910. The memory 906 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0242] The program 910 can be specifically used to enable the processor 902 to execute any resource monitoring method proposed in the above embodiments.

[0243] An embodiment of the present application further provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, an embodiment of the resource monitoring method described above is implemented.

[0244] An embodiment of the present application provides a computer program, which can be executed by a processor to implement an embodiment of the above-mentioned resource monitoring method.

[0245] An embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, an embodiment of the resource monitoring method described above is implemented.

[0246] In several embodiments provided in the present application, if any function is implemented in the form of a software function module / unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, part or all of the technical solution of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, server or other electronic device) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (RandomAccess Memory, RAM), disk or optical disk and other media that can store computer program code.

[0247] The algorithm or display provided here are not inherently related to any specific computer, virtual system or other equipment. Various general systems can also be used together with the teaching based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the present application embodiment is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present application described here, and the above description of specific languages ​​is to disclose the best mode of implementation of the present application.

[0248] It should be noted that the above embodiments illustrate the present application rather than limit the present application, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be constructed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of multiple such elements. The present application may be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In the claims that list several devices, several units or modules in these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

[0249] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A resource monitoring method, characterized in that: include: Get the standard read and write duration information corresponding to multiple sensor resources; Determine the number of groups M for grouping the plurality of sensor resources according to the standard read / write duration information, where M is a positive integer; Acquire measurement information corresponding to a plurality of the sensor resources; Determine the impact factors corresponding to the plurality of sensor resources according to the measurement information corresponding to the plurality of sensor resources; Performing cluster analysis on the influencing factors corresponding to the plurality of sensor resources according to the grouping number M, dividing the plurality of sensor resources into M sensor resource groups, and determining the influencing factors corresponding to the plurality of sensor resource groups; Determining monitoring periods corresponding to the plurality of sensor resource groups according to the influencing factors corresponding to the plurality of sensor resource groups; The sensor resources in the multiple sensor resource groups are monitored respectively according to the monitoring periods corresponding to the multiple sensor resource groups.

2. The resource monitoring method according to claim 1, characterized in that: The step of determining the number of groups M for grouping the plurality of sensor resources according to the standard read / write duration information includes: Determine total read / write duration information T according to standard read / write duration information corresponding to multiple sensor resources; Determine the maximum standard read / write duration information T from the standard read / write duration information corresponding to the plurality of sensor resources max ; The total read / write duration information T and the maximum standard read / write duration information T max After rounding the quotient between , the number of groups M for grouping the plurality of sensor resources is determined.

3. The resource monitoring method according to claim 2, characterized in that: The method further comprises: Determine the minimum standard read / write duration information T from the standard read / write duration information corresponding to the plurality of sensor resources min ; Before obtaining the measurement information corresponding to the plurality of sensor resources, the method further includes: The total read / write time information T and the minimum standard read / write time information T min After rounding the quotient, determine the steady-state number L; If within the L*T time period, the change in the measurement information of the plurality of sensor resources is less than a preset threshold, the step of obtaining the measurement information corresponding to the plurality of sensor resources is performed.

4. The resource monitoring method according to claim 3, characterized in that: The obtaining of measurement information corresponding to the plurality of sensor resources includes: Get the upper and lower operating temperature values ​​of the processor, network, and hard disk; According to the lower limit of the working temperature and the preset temperature step, the working environment temperature is configured in sequence, and the following steps are performed at multiple working environment temperatures until the working environment temperature is greater than or equal to the upper limit of the working temperature: The processor is unloaded, and the network and the hard disk have no tasks; Respectively obtaining measurement information of the plurality of sensors; gradually pressurizing the processor at a preset processor pressurization step size; Respectively obtaining measurement information of the plurality of sensors; gradually decompressing the processor at a preset processor decompression step size; Respectively obtaining measurement information of the plurality of sensors; The processor is unloaded, and the network and the hard disk have no tasks; Respectively obtaining measurement information of the plurality of sensors; gradually pressurizing the network with a preset network pressurization step size; Respectively obtaining measurement information of the plurality of sensors; gradually decompressing the network with a preset network decompression step size; Respectively obtaining measurement information of the plurality of sensors; The processor is unloaded, and the network and the hard disk have no tasks; Respectively obtaining measurement information of the plurality of sensors; gradually pressurizing the hard disk with a preset hard disk pressurization step length; Respectively obtaining measurement information of the plurality of sensors; gradually decompressing the hard disk with a preset hard disk decompression step length; The measurement information of the plurality of sensors is obtained respectively.

5. The resource monitoring method according to claim 1, characterized in that: The determining, according to the measurement information corresponding to the plurality of sensor resources, the influencing factors corresponding to the plurality of sensor resources comprises: Determining an average value of the measurement information corresponding to the plurality of sensor resources according to the measurement information of the plurality of sensor resources; Determine the variance and standard deviation of the measurement information corresponding to the plurality of sensor resources according to the average value of the measurement information corresponding to the plurality of sensor resources; Standardizing the variance and standard deviation of the measurement information corresponding to the plurality of sensor resources to obtain standardized measurement information corresponding to the plurality of sensor resources; determining a correlation coefficient between the plurality of sensor resources according to the standardized measurement information corresponding to the plurality of sensor resources; Generate a correlation coefficient matrix according to the correlation coefficients between the plurality of sensor resources; generating a plurality of feature vectors corresponding to the sensor resources according to the correlation coefficient matrix; generating a load matrix according to the characteristic vectors corresponding to the plurality of sensor resources; Influencing factors corresponding to a plurality of the sensor resources are generated according to the load matrix.

6. The resource monitoring method according to claim 5, characterized in that: The generating of the impact factors corresponding to the plurality of sensor resources according to the load matrix further includes: rotating the load matrix; Influencing factors corresponding to a plurality of sensor resources are generated according to the rotated load matrix.

7. The resource monitoring method according to claim 6, characterized in that: The determining the monitoring periods corresponding to the plurality of sensor resource groups according to the impact factors corresponding to the plurality of sensor resource groups comprises: Taking the inverses of the impact factors corresponding to the plurality of sensor resource groups respectively, obtaining a minimum inverse value from the plurality of inverses, and taking the minimum inverse value as a unit monitoring period; The monitoring periods corresponding to the multiple sensor resource groups are determined by rounding off the quotients of the reciprocals corresponding to the multiple sensor resource groups and the unit monitoring period.

8. A resource monitoring device, characterized in that: include: Sensor resource acquisition module: used to obtain standard read and write duration information and measurement information corresponding to multiple sensor resources; A grouping number determination module: used to determine the number of groups M for grouping the multiple sensor resources according to the standard read and write duration information corresponding to the multiple sensor resources, where M is a positive integer; An impact factor determination module: configured to determine the impact factors corresponding to the plurality of sensor resources according to the measurement information corresponding to the plurality of sensor resources; A sensor grouping module: configured to perform cluster analysis on the influencing factors corresponding to the plurality of sensor resources according to the grouping number M, divide the plurality of sensor resources into M sensor resource groups, and determine the influencing factors corresponding to the plurality of sensor resource groups; A monitoring period determination module: used to determine the monitoring periods corresponding to the plurality of sensor resource groups according to the influencing factors corresponding to the plurality of sensor resource groups; Resource monitoring module: used for monitoring the sensor resources in the plurality of sensor resource groups respectively according to the monitoring periods corresponding to the plurality of sensor resource groups.

9. A resource monitoring device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store executable instructions, and the executable instructions enable the processor to perform operations of the resource monitoring method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one executable instruction. When the executable instruction is executed on the resource monitoring device, the resource monitoring device executes the operation of the resource monitoring method according to any one of claims 1 to 7.

Images