Abstract

Software Defined Networking (SDN) is an emerging phenomenon in computer networking. SDN splits control element and forwarding element of a network. It has characteristics such as centralized management, programmability, more security, granularity and reliability. Therefore SDN is suitable for data center, cloud computing and carrier networks. As reliability and fault tolerance are main concerns, these networks inherently add multiple paths between any two nodes in the network, thus creating opportunities for potential gain in network capacity (throughput). Back-pressure algorithm which uses all possible paths between the source and destination nodes is proven as a throughput optimal routing and scheduling algorithm. However, it exhibits excessively high delays and jitter. In order to realize the key benefit of gain in network capacity, in this work we propose to use back-pressure routing in software-defined networks. However, we address the problem of high delays and jitter associated with back-pressure algorithm by using centralized control and monitoring feature of SDN. The modified backpressure algorithm is evaluated through extensive simulations and observed a significant reduction in delays while retaining throughput optimality of the basic back-pressure algorithm.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call