SDN-based traffic scheduling scheme for energy storage data center based on differential algorithm and congestion degree

In recent years, energy storage power stations gradually merge into data centers. This paper studies the traffic scheduling problem of the energy storage data center network to improve network performance. A traffic scheduling scheme based on differential algorithms and congestion is proposed with the idea of separating the control plane from the data plane of the Software Defined Network. This scheme is divided into two parts: the introduction and improvement of the differential algorithm for unified scheduling of multiple traffic with the same sources and sinks; the concept of congestion degree is proposed for the remaining single traffic to be processed. This paper uses the Mininet network simulation platform and Ryu controller to perform simulation verification in four aspects of standardized throughput, link utilization, load balancing, and link bandwidth utilization. The comparative experiments show that the strategy proposed in this paper can greatly improve the network performance in the above aspects.