Adaptive evaluation method for data center network equipment health

Abstract

The invention discloses an adaptive evaluation method for the health degree of data center network equipment. The steps of the method are as follows: 1) establishing a health degree model of the equipment, and taking the weighted sum of each performance parameter of the equipment and the weight value corresponding to the performance parameter as the health degree 2) Obtain the network topology description information and collect the status information of all network interfaces of each device according to the position of the device in the network, and calculate the performance of the device's message receiving status and disconnection status parameters; 3) Set parameter weights and adjust them according to the probability of occurrence of performance parameters, calculate the health index of the device according to the health model, and then adjust the parameter weights according to the calculated health index until the health index is adjusted to the preset range Inside. The invention has the advantages of simple implementation method, effective and comprehensive evaluation of equipment health degree in a large-scale network, and self-adaptive and dynamic adjustment according to the network structure.

Description

technical field [0001] The invention relates to the technical field of network health evaluation, in particular to an adaptive evaluation method for the health degree of network equipment in a data center. Background technique [0002] Large-scale data centers usually use a large number of core switches, aggregation switches and access switches to form a hierarchical high-speed Ethernet, and are connected to the data center computing server cluster and large-capacity storage system respectively, responsible for providing services between computing servers and computing servers. It provides high-speed information transmission support with the storage system, so the availability and transmission performance of the data center network are important factors affecting the external service capabilities of the data center. In the actual operation and maintenance of the data center network, due to the large scale of equipment and complex link structure in the large-scale network, th...

AI Technical Summary

Problems solved by technology

The advantage of this method is that by dividing complex network problems into simple local modules, it can quickly evaluate the overall status of the data center network equipment; the disadvantage is that it does not consider the hierarchical structure of network equipment deployment, and cannot perform dynamic analysis according to the hierarchical structure of network equipment. adjustment

[0008] 5) By classifying and analyzing various factors in the network such as hardware, service, and transmission one by one, it is possible to judge whether the network is faulty. Health status is graded and judged

Images