Service system running state intelligent research and judgment system based on log, flow and service access

Abstract

The invention discloses a service system running state intelligent research and judgment system based on logs, flow and service access, which reconstructs a comprehensive network management system technical architecture, and comprises the following steps of: constructing a distributed data storage platform by utilizing a plurality of databases; the kafka distributed deployment environment is utilized to realize unified collection and centralized pushing of data in the distributed data storage platform, and the data is sent to the comprehensive network management system; the comprehensive network management system constructs an application system port monitoring and physical storage monitoring system and a display system; and summarizing and integrating key service system operation logs, middleware operation logs and service process operation information according to the acquired data, and mining association relationships between different log modes and service system faults and low performance phenomena by utilizing a big data intelligent analysis method to form a key service system operation state intelligent research and judgment model. The method can effectively improve the guarantee capability of an information network and key services, and reduces the operation risk of a service system.

Description

technical field [0001] The invention relates to the field of data monitoring, in particular to an intelligent judgment system for the operating status of a business system based on logs, traffic and business access. Background technique [0002] First of all, the integrated network management system designed and developed based on the traditional architecture has a sharp increase in the number of information and communication network resource devices, which puts forward higher requirements for network device management and operation reliability, and performance bottlenecks in data collection, query and processing have appeared; and The application deployment, upgrade and expansion are relatively complicated, and the adaptability to the changes of the monitored objects is poor. [0003] Secondly, with the rapid development of the power business, the normal and stable operation of the power core business system is the key to the normal operation of the company's business. Thi...

AI Technical Summary

Problems solved by technology

[0005] 1. In the case of a large number of alarms, the collection judgment can only be made by pre-configuring the collection strategy. Alarm collection cannot be realized for devices without policies configured, that is, the dynamic alarm collection of the entire network cannot be realized, which seriously affects monitoring personnel. Research and judgment on alarms;

[0006] 2. In business system monitoring, it is impossible to intelligently diagnose and analyze component monitoring, process abnormality monitoring, and interface throughput monitoring for abnormal conditions and fault location

The fault prediction time Δtl of this type of method is short, but the fault occurrence time window Δtp is also very short, so the accuracy is high, but the difficulty is that the business import to build an accurate model is usually insufficient, and has a high construction cost. Therefore, this method has not been widely used

Images