CN-122001785-A - Processing method, device, equipment and storage medium of network node

Abstract

The application provides a processing method, a device, equipment and a storage medium of a network node, wherein the method comprises the steps of obtaining performance data of the network node; the method comprises the steps of identifying performance data to obtain a plurality of performance burr points, determining the duration time of each performance burr point, determining a target duration time interval to which the duration time belongs in a plurality of mutually non-overlapping preset duration time intervals, determining the severity level corresponding to the target duration time interval as the target level corresponding to the performance burr point, and executing corresponding maintenance processing on the network node according to the target level corresponding to each performance burr point. By the scheme, the system can respond differently to the anomalies of different degrees, and timely intervention on serious faults is ensured while excessive consumption of operation and maintenance resources on slight problems is avoided, so that the resource allocation efficiency is optimized as a whole and the system stability is enhanced.

Inventors

• LI ZELONG

Assignees

• 北京金山云网络技术有限公司

Dates

Publication Date

20260508

Application Date

20260121

Claims (11)

1. 1. A method of processing a network node, the method comprising: Acquiring performance data of a network node; performing identification processing on the performance data to obtain a plurality of performance burr points; Determining, for each performance burr point, a duration of the performance burr point; Determining a target duration time interval to which the duration time belongs in a plurality of non-overlapping preset duration time intervals, and determining a severity level corresponding to the target duration time interval as a target level corresponding to the performance burr point; and executing corresponding maintenance processing on the network node according to the target grade corresponding to each performance burr point.
2. 2. The method according to claim 1, wherein the performing the corresponding maintenance process on the network node according to the target level corresponding to each performance burr point includes: root cause analysis is carried out according to the target grade corresponding to each performance burr point, and an analysis result is obtained; Generating a maintenance decision according to the analysis result; and executing corresponding maintenance processing on the network node according to the maintenance decision.
3. 3. The method of claim 2, wherein the performing root cause analysis according to the target level corresponding to each performance burr point to obtain an analysis result includes: Aiming at each performance burr point, according to the target grade corresponding to the performance burr point, the occurrence time and the node attribute of the performance burr point, the system event log and the resource state index in the same period are searched in a correlation way; and generating a relevance analysis report according to the system event log and the resource state index, and taking the relevance analysis report as the analysis result.
4. 4. The method of claim 2, wherein the performing root cause analysis according to the target level corresponding to each performance burr point to obtain an analysis result includes: aiming at each performance burr point, evaluating the influence range of the performance burr point on a business service chain according to the target grade of the performance burr point; And generating an influence range evaluation report based on the influence range, and taking the influence range evaluation report as the analysis result.
5. 5. The method of claim 2, wherein the performing root cause analysis according to the target level corresponding to each performance burr point to obtain an analysis result includes: Performing cluster analysis on a plurality of performance burr points with the same target level in a history period to obtain a time distribution mode; And generating a trend prediction report based on the time distribution mode, and taking the trend prediction report as the analysis result.
6. 6. The method according to claim 2, wherein the method further comprises: Binding the analysis result with the maintenance decision to form a strategy entry in a processing knowledge base; performing matching search in the processing knowledge base based on the attribute of the new performance burr point under the condition that the new performance burr point is identified; And pushing the maintenance decision recorded in the strategy item to a maintenance interface under the condition that the strategy item is matched with the corresponding strategy item.
7. 7. The method according to claim 1, wherein the performing the corresponding maintenance process on the network node according to the target level corresponding to each performance burr point includes: aiming at each performance burr point, generating alarm information of corresponding severity according to a target grade corresponding to the performance burr point; And distributing the alarm information to the corresponding maintenance terminal based on a preset notification routing strategy.
8. 8. The method of claim 7, wherein the method further comprises: Counting the number of alarm information of a preset level generated in a first preset time; generating upgrade alarm information under the condition that the quantity reaches a first threshold value; And sending the upgrade alarm information to a preassigned upgrade notification terminal.
9. 9. A processing apparatus of a network node, the apparatus comprising: The acquisition module is used for acquiring the performance data of the network node; The identification module is used for carrying out identification processing on the performance data to obtain a plurality of performance burr points; A first determining module for determining, for each performance burr point, a duration of the performance burr point; The second determining module is used for determining a target duration time interval to which the duration time belongs in a plurality of preset duration time intervals which are not overlapped, and determining a severity level corresponding to the target duration time interval as a target level corresponding to the performance burr point; and the processing module is used for executing corresponding maintenance processing on the network node according to the target grade corresponding to each performance burr point.
10. 10. An electronic device comprising a processor and a memory, the processor being configured to execute a processing program of a network node stored in the memory, to implement the method of processing a network node according to any one of claims 1-8.
11. 11. A storage medium storing one or more programs executable by one or more processors to implement the method of processing a network node of any of claims 1-8.

Description

Processing method, device, equipment and storage medium of network node Technical Field The present application relates to the field of internet technologies, and in particular, to a method, an apparatus, a device, and a storage medium for processing a network node. Background In the field of information technology, large-scale distributed systems are made up of numerous network nodes. These nodes generate performance data during operation, including central processing unit (Central Processing Unit, CPU) utilization, memory usage, request latency, etc. Performance glitches, i.e., transient abnormal fluctuations in these indicators, are key factors affecting system stability. The prior art generally employs a fixed threshold based monitoring scheme. The scheme presets a static threshold for each performance index, and when the monitoring system detects that the performance data of a certain node exceeds the threshold, a unified alarm signal is triggered to inform operation and maintenance personnel to check. However, the existing scheme cannot accurately evaluate the severity of the performance abnormality, and accordingly, differential operation and maintenance response is performed, so that the operation and maintenance resource allocation efficiency is low, and the system stability is difficult to effectively guarantee. Disclosure of Invention The application provides a processing method, a device, equipment and a storage medium of a network node, which are used for solving the problems that the prior art cannot accurately evaluate the severity of abnormal performance, and differentiated operation and maintenance response is carried out according to the severity, so that the distribution efficiency of operation and maintenance resources is low, and the stability of a system is difficult to effectively guarantee. In a first aspect, the present application provides a method for processing a network node, including: Acquiring performance data of a network node; performing identification processing on the performance data to obtain a plurality of performance burr points; Determining, for each performance burr point, a duration of the performance burr point; Determining a target duration time interval to which the duration time belongs in a plurality of non-overlapping preset duration time intervals, and determining a severity level corresponding to the target duration time interval as a target level corresponding to the performance burr point; and executing corresponding maintenance processing on the network node according to the target grade corresponding to each performance burr point. In one possible implementation manner, the performing, according to the target level corresponding to each performance burr point, a corresponding maintenance process on the network node includes: root cause analysis is carried out according to the target grade corresponding to each performance burr point, and an analysis result is obtained; Generating a maintenance decision according to the analysis result; and executing corresponding maintenance processing on the network node according to the maintenance decision. In one possible implementation manner, the performing root cause analysis according to the target level corresponding to each performance burr point to obtain an analysis result includes: Aiming at each performance burr point, according to the target grade corresponding to the performance burr point, the occurrence time and the node attribute of the performance burr point, the system event log and the resource state index in the same period are searched in a correlation way; and generating a relevance analysis report according to the system event log and the resource state index, and taking the relevance analysis report as the analysis result. In one possible implementation manner, the performing root cause analysis according to the target level corresponding to each performance burr point to obtain an analysis result includes: aiming at each performance burr point, evaluating the influence range of the performance burr point on a business service chain according to the target grade of the performance burr point; And generating an influence range evaluation report based on the influence range, and taking the influence range evaluation report as the analysis result. In one possible implementation manner, the performing root cause analysis according to the target level corresponding to each performance burr point to obtain an analysis result includes: Performing cluster analysis on a plurality of performance burr points with the same target level in a history period to obtain a time distribution mode; And generating a trend prediction report based on the time distribution mode, and taking the trend prediction report as the analysis result. In one possible embodiment, the method further comprises: Binding the analysis result with the maintenance decision to form a strategy entry in a processing knowledge base; perfor