Load balancing using multipath routing in network functions virtualization

Network Functions Virtualization (NFV) is a new approach in architectural model of communication networks, where network functions are softwarized and can be deployed dynamically on most different spots in the network, replacing the specialized equipments located at predetermined locations. NFV comes out as an important transformation trend in the recent development of communication networks due to advantages such as the ability to highly customize the network scale and minimize the investment costs. In this research, we analyze and solve the load balancing problem using multipath routing in NFV to optimize network performance and improve responsive ability of a network system based on NFV according to the dynamic changes of user demands. We propose the FAST-RACE algorithm which has ability to give a weight system on the links of the NFV infrastructure in order to equally distribute load among the several shortest paths and minimize the network delay of data flows in the network. The experiment results show that FAST-RACE declines the average delay of data flows about 26% and could support approximately a 14% increase in the number of service demands in comparison to the load balancing solution using hop-count. Specially, the effective performance of FAST-RACE is not subject to a number of service demands and a number of nodes in the NFV infrastructure.

[1]  Nick McKeown,et al.  OpenFlow: enabling innovation in campus networks , 2008, CCRV.

[2]  Serge Fdida,et al.  Pricing in Information-Centric Network interconnection , 2013, 2013 IFIP Networking Conference.

[3]  Mark Handley,et al.  Improving datacenter performance and robustness with multipath TCP , 2011, SIGCOMM 2011.

[4]  John Moy,et al.  OSPF Version 2 , 1998, RFC.

[5]  Meral Shirazipour,et al.  Optical service chaining for network function virtualization , 2015, IEEE Communications Magazine.

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

[7]  Mikkel Thorup,et al.  Internet traffic engineering by optimizing OSPF weights , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[8]  K. K. Ramakrishnan,et al.  NetVM: High Performance and Flexible Networking Using Virtualization on Commodity Platforms , 2014, IEEE Transactions on Network and Service Management.

[9]  Mikkel Thorup,et al.  Increasing Internet Capacity Using Local Search , 2004, Comput. Optim. Appl..

[10]  Byrav Ramamurthy,et al.  Network Innovation using OpenFlow: A Survey , 2014, IEEE Communications Surveys & Tutorials.

[11]  Jianping Wang,et al.  Optimization Models for Congestion Mitigation in Virtual Networks , 2014, 2014 IEEE 22nd International Conference on Network Protocols.

[12]  Anthony E. Krzesinski,et al.  Path selection and bandwidth allocation in MPLS networks , 2003, Perform. Evaluation.

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

[14]  Serge Fdida,et al.  DTN support for news dissemination in an urban area , 2012, Comput. Networks.

[15]  Alberto Leon-Garcia,et al.  Routing Algorithms for Network Function Virtualization Enabled Multicast Topology on SDN , 2015, IEEE Transactions on Network and Service Management.

[16]  Roch Guérin,et al.  Achieving near-optimal traffic engineering solutions for current OSPF/IS-IS networks , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[17]  Albert G. Greenberg,et al.  Ananta: cloud scale load balancing , 2013, SIGCOMM.

[18]  Ross W. Callon,et al.  Use of OSI IS-IS for routing in TCP/IP and dual environments , 1990, RFC.