การดุลภาระของคอนโทรลเลอร์โดยตระหนักถึงการบริโภคพลังงานในซอฟต์แวร์ดีฟายด์เน็ตเวิร์กกิงด้วยการย้ายสวิตช์

Abstract

Software-Defined Networking (SDN) decouples the control and data planes, enhancing both network management and programmability. SDN switches in the data plane communicate with controllers, generating controller loads. However, dynamic network load changes can result in controller overloads. Some switches may need to be migrated to other controllers to solve the problem. With the growing focus on energy efficiency in networking, minimizing energy consumption from communications between switches and controllers becomes important. Thus, a load distribution strategy that also accounts for energy consumption is needed. This thesis presents the Energy-aware Controller Load Distribution (ECLD) algorithm. It addresses the controller overload issue and aims to reduce the overall energy consumption associated with switch loads. It considers hop counts and energy costs, as well as the effect where a switch migration causes others to migrate to further controllers, increasing energy consumption. Evaluation results indicate that the ECLD algorithm achieves lower energy consumption compared to existing algorithms.