Performance Evaluation of Fault Tolerant Routing Algorithm in Data Center Networks

Abstract

Nowdays, the vigorous development of cloud computing technology has brought great changes to the development of the whole information industry. The traditional data center network topology construction method and the operation mechanism of the network layer control plane are solidified, which have been difficult to meet the increasing demand for high performance and high cost performance under the new situation. Researchers map the topology of the data center network to an undirected graph, and use graph algorithms to implement fault-tolerant routing in complex networks. However, the shortest path algorithm of some early graphs cannot be applied to all topological graphs. In this paper, we analyze the advantages and disadvantages of related algorithms, and propose a fault-tolerant routing algorithm for all networks, which can dynamically call different algorithms according to the current number of nodes and links. The experimental results show that it can effectively improve the accuracy and fault tolerance of the algorithm and reduce the consumption of time and memory.