A Scalable and Distributed Approach for NFV Service Chain Cost Minimization

Network function virtualization (NFV) represents the latest technology advancement in network service provisioning. Traditional hardware middleboxes are replaced by software programs running on industry standard servers and virtual machines, for service agility, flexibility, and cost reduction. NFV users are provisioned with service chains composed of virtual network functions (VNFs). A fundamental problem in NFV service chain provisioning is to satisfy user demands with minimum system-wide cost. We jointly consider two types of cost in this work: nodal resource cost and link delay cost, and formulate the service chain provisioning problem using nonlinear optimization. Through the method of auxiliary variables, we transform the optimization problem into its separable form, and then apply the alternating direction method of multipliers (ADMM) to design scalable and fully distributed solutions. Through simulation studies, we verify the convergence and efficacy of our distributed algorithm design.

[1]  Seungjoon Lee,et al.  Network function virtualization: Challenges and opportunities for innovations , 2015, IEEE Communications Magazine.

[2]  Zheng Xu,et al.  Training Neural Networks Without Gradients: A Scalable ADMM Approach , 2016, ICML.

[3]  황정연,et al.  클라우드 컴퓨팅 기반의 NFV Use Cases , 2016 .

[4]  Joseph Naor,et al.  Near optimal placement of virtual network functions , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[5]  Fulvio Risso,et al.  Research Directions in Network Service Chaining , 2013, 2013 IEEE SDN for Future Networks and Services (SDN4FNS).

[6]  S. Shenker.  Fundamental Design Issues for the Future Internet , 1995 .

[7]  Zongpeng Li,et al.  An efficient auction mechanism for service chains in the NFV market , 2016, IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications.

[8]  Filip De Turck,et al.  VNF-P: A model for efficient placement of virtualized network functions , 2014, 10th International Conference on Network and Service Management (CNSM) and Workshop.

[9]  Patrick Pérez,et al.  Distributed Non-convex ADMM-based inference in large-scale random fields , 2014, BMVC.

[10]  Baochun Li,et al.  Joint request mapping and response routing for geo-distributed cloud services , 2013, 2013 Proceedings IEEE INFOCOM.

[11]  Abdallah Shami,et al.  NFV: state of the art, challenges, and implementation in next generation mobile networks (vEPC) , 2014, IEEE Network.

[12]  Patrick Pérez,et al.  Distributed Non-Convex ADMM-inference in Large-scale Random Fields , 2014 .

[13]  B. Mercier,et al.  A dual algorithm for the solution of nonlinear variational problems via finite element approximation , 1976 .

[14]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..