An Effective Routing Mechanism Based on Fuzzy Logic for Software-Defined Data Center Networks

Abstract

In order to solve the problem of low throughput and poor performance in data center network, this paper proposes a traffic scheduling algorithm exploiting fuzzy logic inference based on Software Defined Networking (SDN). The algorithm uses the characteristics of separation of control and forwarding in SDN architecture to choose paths for data flows by the centralized control of controller. It calculates the candidate path set of accessible paths between the source host and the destination host, then comprehensively considers path hops and bandwidth utilization, evaluates the candidate path using fuzzy logic model, and selects the optimal path. The experimental results show that compared with ECMP, Hedera and FSEM, the proposed algorithm improves the network throughput and load balancing, thus improving the overall network performance of the data center.