MOSC: a method to assign the outsourcing of service function chain across multiple clouds

Research paper by Huan Chen, Xiong Wang; Yangming Zhao; Tongyu Song; Yang Wang; Shizhong Xu; Lemin Li

Indexed on: 04 Apr '18Published on: 04 Feb '18Published in: Computer Networks


Publication date: 14 March 2018 Source:Computer Networks, Volume 133 Author(s): Huan Chen, Xiong Wang, Yangming Zhao, Tongyu Song, Yang Wang, Shizhong Xu, Lemin Li As Network Function Virtualization (NFV) becomes reality and cloud computing offers a scalable pay-as-you-go charging model, more network operators would like to outsource their Service Function Chains (SFC) to the public clouds in order to reduce the operational cost. Unfortunately, challenges of Quality of Service guarantee still exist while minimizing the operational cost with outsourcing SFC to public clouds. In this paper, we investigate this problem when there are a large number of candidate cloud providers with the consideration of diverse pricing schemes of network functions, additional latency caused by public network, and the relationship between the Virtual Network Function (VNF) performance and its cost. Compared with our previous conference version, we design D-MOSC, an improved deviation based heuristic algorithm to assign the Outsourcing of SFC across multiple clouds based on Hidden Markov Model (HMM). The extensive simulations show that MOSC saves up to 79.2% cost compared with that of deploying network functions in the local network. MOSC also achieves up to 50.7% cost savings compared with the result of the first-fit based optimization algorithm. Compared with the greedy version, D-MOSC achieves up to 26.7% cost savings with the guarantee of latency requirements.