Hierarchical policy based data center network availability assessment method and assessment apparatus

Abstract

The present invention relates to a hierarchical policy based data center network availability assessment method and assessment apparatus. The method includes a first step of defining topological information of a data center network; a second step of dynamically collecting performance parameter data of network devices in the data center network; a third step of implementing performance state assessment of a network switch device; a fourth step of setting network layers; a fifth step of initializing priorities of the network devices; and a sixth step of performing performance state assessment on the whole network, that is, based on the availability assessment quantized value of each network device according to the set hierarchical policy, analyzing by referring to a network topological structure so as to obtain different impacts of the network devices at all topological positions on the network, and obtain a running state quantized value of the whole network. The assessment apparatus is used for implementing the foregoing assessment method. The method has the advantages of wide application range, high assessment efficiency, good assessment effect and so on.

Description

technical field [0001] The present invention mainly relates to the technical field of data center network, in particular to a data center network usability evaluation method and evaluation device based on hierarchical strategy. Background technique [0002] In order to achieve faster and more efficient data transmission in large-scale data centers, a large number of core switches, aggregation switches and access switches are used to build a layered high-speed transmission Ethernet to realize the connection between computing node server clusters and large-capacity storage systems in the data center. It is mainly responsible for providing high-speed and stable information transmission support for the interaction between computing node servers and data flow between computing node servers and storage systems. Therefore, the availability and transmission performance of the data center network are important factors that affect the external service capabilities of the data center. ...

AI Technical Summary

Problems solved by technology

Disadvantages: 1) In the calculation process of entropy weight, the key parameters need to be determined by experience; 2) The classification method is only suitable for specific business networks, not suitable for complex data center networks

[0013] (1) Due to the large volatility of the usability evaluation value parameters, the network evaluation value may be inaccurate and cannot reflect the real network situation

[0014] (2) The sample values ​​of the collected usability evaluation parameters are relatively large, which causes the normalized results to deviate from the normal operating values, and also causes large errors in the evaluation results

[0015] (3) The proposed grading method is only applicable to specific business networks, and the network availability evaluation method is very complicated

Images