Research On Network Energy-saving Technology Based On OpenFlow

With the rapid development of mobile Internet,cloud computing,big data and other internet technologies,the scale of network continues to expand and the new network architecture also appear endless.To enhance network robustness and respond to network burst traffic,the problem of excess resource supply and redundant design in traditional network as well as the mismatch between network equipment energy consumption and traffic volume result in low utilization of network energy,wasting large of energy.In order to solve the problem of low utilization in network energy,the basic idea is to confine the network traffic into the sub-topology then close the idle network links or equipment.But in the traditional network,the issue of difficulties in obtaining real-time traffic network and the network of energy-saving operation bringing a lot of network overhead and so on result in that energy efficiency and QoS can not meet at the same time.OpenFlow as the key technology to separate the network control function from data forwarding function,has become the universal southbound interface protocol of software defined networking(SDN).OpenFlow controller greatly enhances the flexibility of the network control,which provides a new way to achieve network energy-saving.Based on the certain regularity of network traffic distribution,using the OpenFlow technology to sleep and wake up the equipment,obtaining real-time traffic and changing the traffic route,an energy-saving algorithm based on real-time traffic is proposed.Firstly,this algorithm uses link utilization as QoS constraint,and the link whose link utilization exceed the threshold of congestion constraints need to do load-balancing.Then In order to cope with bursty traffic in the network,an improved algorithm is proposed which can update threshold to match the network load.Finally,the RTT-ES energy-saving system is realized based on the enterprise-class network controller Floodlight and Mininet simulation platform,and make a detail analysis and assessment to the RTT-ES algorithm.