CN-115328740-B - Power supply state monitoring method and device, electronic equipment and storage medium

CN115328740BCN 115328740 BCN115328740 BCN 115328740BCN-115328740-B

Abstract

The application discloses a power state monitoring method and device, electronic equipment and a storage medium, wherein the method comprises the steps of obtaining equipment information of each target server and corresponding information acquisition script information thereof; the method comprises the steps of collecting power supply information of each target server in a preset time period based on equipment information of each target server and corresponding information collecting script information of the equipment information, comparing and analyzing circuit currents of each power supply in the power supply information of each target server for each target server to obtain a comparison analysis result of the target server, judging whether power supply of the target server is abnormal or not based on the comparison analysis result of the target server, extracting target characteristic information of the target server from the power supply information of the target server if the abnormality is judged to exist, inputting the target characteristic information of the target server into a pre-trained prediction model to obtain a power supply failure prediction result of the target server, and feeding back the power supply failure prediction result of the target server.

Inventors

ZHANG YU
LIU QIAN
CHEN LONGHUI

Assignees

中国建设银行股份有限公司

Dates

Publication Date: 20260512
Application Date: 20220906

Claims (8)

1. A power state monitoring method, comprising: Acquiring equipment information of each target server and information acquisition script information corresponding to each target server; Acquiring power supply information of each target server in a preset time period based on equipment information of each target server and the information acquisition script information corresponding to each target server; Respectively carrying out comparison analysis on the circuit currents of all power supplies of the target servers in the power supply information of the target servers aiming at all the target servers to obtain comparison analysis results of the target servers; Judging whether the power supply of the target server is abnormal or not based on the comparison analysis result of the target server; If the power supply of the target server is judged to be abnormal, extracting target characteristic information of the target server from the power supply information of the target server; Inputting the target characteristic information of the target server into a pre-trained prediction model, and processing the target characteristic information of the target server through the prediction model to obtain a power failure prediction result of the target server; feeding back a power failure prediction result of the target server; The comparing and analyzing the circuit current of each power supply of the target server in the power supply information of the target server to obtain a comparing and analyzing result of the target server includes: calculating, for each target server, a current variance corresponding to each power supply by using a circuit current of each power supply of the target server in the power supply information of the target server; calculating the difference value of the current variances corresponding to the power supplies to obtain a current deviation difference value; Performing correlation analysis on circuit currents of all power supplies of the target server in the power supply information of the target server to obtain correlation degrees among all power supplies; And determining the current deviation difference value and the correlation degree among the power supplies as the comparison analysis result.
2. The method according to claim 1, wherein the obtaining the device information of each target server and the information collection script information corresponding to each target server includes: When a batch acquisition task of each target server is initiated, acquiring equipment information of each target server from a back-end task configuration table according to the identification of the batch acquisition task; And according to the server model in the equipment information of each target server, matching the information acquisition script information corresponding to each target server from the back-end task configuration table.
3. The method of claim 1, wherein the determining whether the power source of the target server is abnormal based on the comparison analysis result of the target server comprises: Judging whether the current deviation difference value in the comparison analysis result of the target server is larger than a preset difference value or whether the correlation degree between any two power supplies in the comparison analysis result of the target server is smaller than a preset degree threshold; If the current deviation difference value in the comparison analysis result of the target server is larger than the preset difference value or the correlation degree between any two power supplies in the comparison analysis result of the target server is smaller than the preset degree threshold value, determining that the power supply of the target server is abnormal.
4. The method of claim 1, wherein after feeding back the power failure prediction result of the target server, further comprising: And if the power failure prediction result of the target server indicates that the power source of the target server has failure risk, carrying out risk prompt on the target server.
5. A power state monitoring device, comprising: The information acquisition unit is used for acquiring equipment information of each target server and information acquisition script information corresponding to each target server; The information acquisition unit is used for acquiring power supply information of each target server in a preset time period based on equipment information of each target server and the information acquisition script information corresponding to each target server; the analysis unit is used for comparing and analyzing the circuit current of each power supply of the target server in the power supply information of the target server aiming at each target server respectively to obtain a comparison and analysis result of the target server; The judging unit is used for judging whether the power supply of the target server is abnormal or not based on the comparison analysis result of the target server; the feature extraction unit is used for extracting target feature information of the target server from the power information of the target server when the power of the target server is judged to be abnormal; The prediction unit is used for inputting the target characteristic information of the target server into a pre-trained prediction model, and processing the target characteristic information of the target server through the prediction model to obtain a power failure prediction result of the target server; The feedback unit is used for feeding back the power failure prediction result of the target server; wherein the analysis unit comprises: The variance analysis unit is used for calculating current variances corresponding to the power supplies by utilizing the circuit currents of the power supplies of the target servers in the power supply information of the target servers for the target servers respectively; The difference value calculation unit is used for calculating the difference value of the current variances corresponding to the power supplies to obtain a current deviation difference value; the correlation degree determining unit is used for carrying out correlation analysis on the circuit currents of the power supplies of the target server in the power supply information of the target server to obtain the correlation degree among the power supplies; And the analysis result determining unit is used for determining the current deviation difference value and the correlation degree among the power supplies as the comparison analysis result.
6. The apparatus according to claim 5, wherein the information acquisition unit includes: The device information acquisition unit is used for acquiring the device information of each target server from a back-end task configuration table according to the identification of the batch acquisition task when the batch acquisition task of each target server is initiated; And the matching unit is used for matching the information acquisition script information corresponding to each target server from the back-end task configuration table according to the server model in the equipment information of each target server.
7. An electronic device, comprising: A memory and a processor; wherein the memory is used for storing programs; The processor is configured to execute the program, and the program is specifically configured to implement the power state monitoring method according to any one of claims 1 to 4 when executed.
8. A computer storage medium storing a computer program which, when executed, is adapted to implement the power state monitoring method of any one of claims 1 to 4.

Description

Power supply state monitoring method and device, electronic equipment and storage medium Technical Field The present application relates to the field of device state monitoring technologies, and in particular, to a method and apparatus for monitoring a power state, an electronic device, and a storage medium. Background Among the components of the server, the power supply is responsible for supplying energy to the components, and plays a very important role. While data centers with tens of thousands of servers typically require 24 hours to run without interruption, the power requirements are higher. If the power failure of the server affects the power supply of the system, the abnormal power-down of the system can be caused, and the failure of the system can affect the integrity of data, so that great loss is caused, and therefore, the running state of the power supply needs to be effectively monitored, so that corresponding maintenance can be performed in time. The current monitoring of the power state is mainly to collect power consumption data of a server, then provide a chart and other modes for operation and maintenance personnel, and monitor whether the power supply is abnormal or not and whether the power supply is abnormal or not through the provided data by the operation and maintenance personnel. Or acquiring several characteristic information of the CPU in the server, and obtaining the health state of the CPU by using a classification algorithm, so as to analyze the state of the power supply. However, by means of monitoring operation and maintenance personnel based on the collected data, abnormal conditions of the power supply cannot be monitored particularly timely, and the comparison depends on personal level, so that the accuracy of monitoring cannot be guaranteed. The state of the CPU is analyzed through a plurality of characteristic information of the CPU, the accuracy is not high, and the power state is required to be further analyzed, so that the monitoring accuracy cannot be ensured. Disclosure of Invention Based on the defects of the prior art, the application provides a power state monitoring method and device, electronic equipment and a storage medium, so as to solve the problem that the accuracy of a monitoring result cannot be effectively ensured in the prior art. In order to achieve the above object, the present application provides the following technical solutions: the first aspect of the present application provides a power state monitoring method, including: Acquiring equipment information of each target server and information acquisition script information corresponding to each target server; Acquiring power supply information of each target server in a preset time period based on equipment information of each target server and the information acquisition script information corresponding to each target server; Respectively carrying out comparison analysis on the circuit currents of all power supplies of the target servers in the power supply information of the target servers aiming at all the target servers to obtain comparison analysis results of the target servers; Judging whether the power supply of the target server is abnormal or not based on the comparison analysis result of the target server; If the power supply of the target server is judged to be abnormal, extracting target characteristic information of the target server from the power supply information of the target server; Inputting the target characteristic information of the target server into a pre-trained prediction model, and processing the target characteristic information of the target server through the prediction model to obtain a power failure prediction result of the target server; And feeding back a power failure prediction result of the target server. Optionally, in the above power state monitoring method, the acquiring device information of each target server and information acquisition script information corresponding to each target server includes: When a batch acquisition task of each target server is initiated, acquiring equipment information of each target server from a back-end task configuration table according to the identification of the batch acquisition task; And according to the server model in the equipment information of each target server, matching the information acquisition script information corresponding to each target server from the back-end task configuration table. Optionally, in the above power state monitoring method, the comparing and analyzing, for each target server, a circuit current of each power source of the target server in power source information of the target server, to obtain a comparison and analysis result of the target server includes: calculating, for each target server, a current variance corresponding to each power supply by using a circuit current of each power supply of the target server in the power supply information of the target server; calculating the di