Dynamic function composition for network service chain: Model and optimization

Current Internet has been unable to adapt to evolving diverse service requirements due to its inflexibility and complexity. Network service chain (NSC) which can promises increased flexibility and cost efficiency for future networks, has received considerable attention lately. However, the realization of NSC in current Internet faces severe challenges. The appearance of Software Defined Networking and Network Function Virtualization makes the realization and deployment of NSC possible. A key feature in NSC-Enabled networks is the capability of dynamic composing service functions into complex services. In this paper, we investigate model and optimization of dynamic function composition problem. We have developed a combinatorial optimization model to describe the resources and components of dynamic function composition, and proposed a distributed algorithm using Markov approximation method. Performance evaluation demonstrates that our algorithm is effective, scalable and cost-efficient, compared to traditional algorithms.