Method for processing network alerting information

Abstract

This invention relates to a handling method for the network alarm information, its execution includes the following steps: network management system collect GSM network resource data and establish a data management platform system; network management system has a abstract and description of the GSM network resource inner property from resource category and resource relation this two dimensions. It establishes resource model through description of GSM network resource association. And it establishes resource object bank for the data in resource data management platform system according to the association of resource model object; the network management system collects and analyses several original warning events, then has a subsumption and management to events and forms consolidation warning information; the network management system merges the related consolidation warning information according to warning relevant analysis rules and resource model, then produces the alarm information. This invention can merge failure warning, remove redundant information, solve problems of alarm deletion, help to determine the failure source and respond to failure treatment quickly.

Description

technical field [0001] The invention relates to a method for processing network alarm information, in particular to a network resource model-based method for processing network alarm information through alarm correlation analysis and distribution. Background technique [0002] In the current GSM traffic network, in order to realize centralized monitoring of network equipment and quickly discover faults on the network, the GSM traffic network management system processes alarm data of wireless network Real-time monitoring of equipment. [0003] However, the existing network management system has the following deficiencies in the alarm management function: [0004] 1. Different manufacturers and types of equipment have different alarm contents, formats, and levels. The alarms of the network management system are collected from different manufacturers and different types of equipment, and are concentrated and presented on an alarm browser after being analyzed and processed by ...

AI Technical Summary

Problems solved by technology

[0005] 2. Alarms are redundant and have no clear relationship

When a managed network element fails, it often does not affect a single network element device, but all associated devices will generate alarms, and the number of alarms will reach a peak in an instant, and a large number of alarms will be displayed on the alarm browser. This kind of alarm is complicated. and disorderly

A simple method currently used to reduce alarms is to first unify and organize the alarms of different types of equipment from different manufacturers, and then filter them to focus on certain types of alarms or alarms with high levels of concern. This method helps to reduce monitoring time. Alarm, but it is impossible to locate the fault based on a comprehensive analysis of network connection conditions and alarm-related conditions

Maintenance personnel can only judge the root cause of the fault based on certain experience and must understand the complex network structure and analyze the relationship between the alarm content and the alarm source in a large number of alarm information, which takes a long time and usually delays Troubleshooting, sometimes the source of the fault cannot be correctly determined

[0006] 3. Missing alarm

Maintenance personnel cannot know that a major failure has occurred in the device through the network management system in time. They can only judge the failure of the device indirectly by observing the alarms of other devices connected to the network based on experience. This also requires a very good understanding of the network structure. It is also difficult to judge

[0007] With the continuous increase of the network scale, the increasing number of network devices and the increasingly complex network structure, communication operators have higher and higher requirements for centralized maintenance of equipment. The above methods cannot solve the problem of large number of alarms and difficulty in rapid failure. positioning problem

[0008] In order to solve the above problems, the alarm correlation analysis method is usually used at present. The existing method is mainly based on the method of alarm correlation rule mining, mining the relationship between events, and realizing it through the event rule engine. The disadvantage is that it does not consider the network topology. Dynamic changes

Another method is to use the event correlation analysis engine and the network topology relationship library to process the formatted information in real time according to the defined rules, generate root-cause alarms or select the most root-cause alarm information. This method can also achieve alarm correlation analysis, but because it uses a relational resource database, its disadvantage is that it is limited by the characteristics of the database, and the efficiency of correlation analysis is low. At the same time, it has poor adaptability to changes in resource information. When resource data relationships change, The application layer function of the event correlation analysis engine must be aware of changes in the information model and make corresponding code modifications

The above two methods are less flexible and scalable

Images