Effective Resource Management in SDN Enabled Data Center Network Based on Traffic Demand

Abstract

Software-defined networking (SDN) changes the way forwarding devices works and makes data center network simplified and efficient to work. Energy is the main driving force behind the activity of the data center network. The network elements that are in low load condition can be shifted in switch- OFF condition after migrating their active traffic on some other element. This traffic analysis-based energy-saving approach will increase the overall utilization of the system. This paper presents a switch- ON/OFF procedure along with the re-routing procedure to achieve the desired activity. The modules are tested over some well-known data center topologies, such as Fat-Tree and Bcube. The simulation study is carried out with the classical SDN controller instances, e.g., OpenDayLight, POX, Beacon, and NOX. The results are compared with some similar energy-efficient data center approaches, such as MNLR DCN and WDM Optical Switched DCN. The proposed approach presents an enhancement of 77%, 37%, and 63% in achieved energy saving, average round trip delay, and average bandwidth utilization, respectively.