Load Balancing in DCN Servers through SDN Machine Learning Algorithm

Abstract

Abstract Software Defined Networking (SDN) manages data traffic in Data Center Network (DCN). SDN improves utilization of large scale network resource and performance of network application. In SDN, load balancing technique optimizes the data flow during transmission through server load deviation after evaluating the network status dynamically. However, load deviation in the network needs optimum server selection and routing path with respect to less time and complexity. In this paper, we propose a M ultiple R egression B ased S earching (MRBS) algorithm for optimum server selection and routing path in DCN. The appropriate server selection during heavy load conditions such as message spikes, different message frequencies and unpredictable traffic patterns are done through regression based analysis and correlation of various server parameters, only after detecting the types of traffic and loads based on bandwidth. The parameters included in the regression modeling are load, response time, bandwidth and server utilization. Moreover, the heuristic algorithm is combined with regression model for efficient path selection. The proposed algorithm reduces the delay and time more than 85% when compared with traditional algorithms due to stochastic gradient decent weights estimation.