Safe and quick detection method of IP network route and route analysis server

Abstract

The invention relates to a safe and quick detection method of an IP network route and a route analysis server. The safe and quick detection method comprises the steps that the route analysis server makes statistics on the numbers of routes derived from all sources AS in a whole network BGP route table in real time, and the inter-class variances of the numbers of the routes deviated from all the sources AS are respectively calculated according to the statistic route numbers and the pre-calculated historical statistical average of the numbers of the routes of all the sources AS; the inter-class variances of the route numbers of all the sources AS are respectively compared with a preset threshold value, whether abnormal fluctuations exist on the routes deviated from all the sources AS or not is judged according to comparing results, and the sources AS of the routes with abnormal fluctuations are determined. According to the safe and quick detection method, the inter-class variances of the numbers of the routes deviated from all the sources AS are calculated to analyze the abnormal fluctuations of the routes, the sources of route fluctuations can be quickly and accurately positioned, statistical features are obvious so that analysis results can be high in accuracy, and the defects of analysis capacity on accurate positioning, statistical features and the like of an existing detection mode are overcome.

Description

technical field [0001] The invention relates to the field of network security, in particular to a fast detection method for IP network routing security and a routing analysis server. Background technique [0002] In an IP network, various factors such as line instability, routing attacks, routing interface hardware problems, and human operations may cause abnormal routing fluctuations in the network. Such fluctuations will cause network instability and lead to A large number of routers need to recalculate routes, resulting in a large waste of bandwidth and CPU resources. [0003] In order to avoid and reduce the impact of abnormal routing fluctuations, it is necessary to detect abnormal routing fluctuations in the network. Existing routing anomaly fluctuation detection methods are judged by recording the curve changes of Internet routing entries, see figure 1 As shown, it specifically includes the following steps: [0004] Step 101, learning to receive Internet routes; ...

AI Technical Summary

Problems solved by technology

[0009] In the above method for detecting abnormal routing fluctuations, the Internet routing entries are not obvious as a statistical feature, and cannot accurately reflect the routing fluctuations, and it is difficult to accurately locate the source of abnormal routing fluctuations

Images