Assessing Profit of Prediction for SDN controllers load balancing

Software-defined networking (SDN) provides programmable control and centralized management in data centers, making it a popular architecture. The large scale of networks has required to propose the geographical distribution logically plane achieve scalability reliability. For solving load imbalance among multiple controllers associated with statically configured plane, switch migration mechanism is proposed admit dynamic balancing. Many studies have been carried out for balancing problem based on mechanism. However, previous focus migrating switches when are overloaded, thereby, wasting time phase resulting high latency. To address these problems, we Assessing Profit Of Prediction (APOP) scheme, load-balancing strategy multiple-controllers overloaded state prediction profit assessment. We introduce Taylor’s formula predict flow change network assess advance, order decrease minimize harmful effects during phase. result simulation experiments shows that our scheme performs effectively reducing cost