Multi-objective optimization method based on probe scheduling and related equipment

Abstract

One or more embodiments of the invention provide a multi-objective optimization method based on probe scheduling and related equipment. The method comprises the following steps: creating a probe coverage rate constraint condition according to different probe coverage indexes; under the condition that the probe coverage rate constraint condition is met, creating a multi-objective optimization modelby taking the load of the probe set and the number of the probe sets as optimization objectives; solving the multi-objective optimization model by using an NSGA2 algorithm to obtain a plurality of satisfactory probe sets; and performing decision analysis on the plurality of satisfactory probe sets by using an analytic hierarchy process (AHP) to obtain an optimal probe set. According to the methodand the related equipment provided by the invention, under the condition of meeting the coverage rate, the probe set with small load and small quantity is preferentially selected, so that the efficiency of measuring the network performance is improved, and meanwhile, the waste of measurement resources is also avoided.

Description

technical field [0001] One or more embodiments of this specification relate to the technical field of computer network performance measurement, and in particular to a multi-objective optimization method based on probe scheduling and related equipment. Background technique [0002] With the rapid development of the Internet, due to the increase of personal information on the Internet and people's dependence on the Internet, the stable operation of the network and security issues have become important factors restricting the development of the Internet. Destructive network failure events will cause application service failures, reduced available bandwidth, and traffic congestion at link ingress, seriously affecting network and service performance. The cause of failure also becomes various. In order to ensure the normal operation of the network, it is necessary to locate network faults in a timely and accurate manner, and establish a complete network detection system to obtain...

AI Technical Summary

Problems solved by technology

[0003] However, due to the limited number of deployed probes and the uneven distribution of probes, if all probes are called to perform measurement tasks, a large number of repeated measurement results may be generated due to the same region and autonomous domain (AS) of the probes. , resulting in a waste of measurement resources, but also increases the overhead of measurement costs, such as additional network traffic overhead and data storage management

[0004] Moreover, with the increasing number of probe measurement tasks, the load of the probe is also increasing, which will gradually reduce the network measurement efficiency of the probe. When the load of the probe is greater than the load threshold, it may cause the probe to crash. machine failure

Images