PROJECT TITLE :

Online Scaling of NFV Service Chains Across Geo-Distributed Datacenters - 2018

ABSTRACT:

Network Function Virtualization (NFV) is an emerging paradigm that turns hardware-dependent implementation of network functions (i.e., middleboxes) into software modules running on virtualized platforms, for vital cost reduction and simple management. Such virtual network functions (VNFs) commonly represent service chains, to supply network services that traffic flows want to travel through. Efficient deployment of VNFs for network service provisioning may be a key to appreciate the NFV goals. Existing efforts on VNF placement mostly pander to offline or one-time placement, ignoring the fundamental, dynamic deployment and scaling need of VNFs to handle practical time-varying traffic volumes. This work investigates dynamic placement of VNF service chains across geo-distributed datacenters to serve flows between dispersed source and destination pairs, for operational price minimization of the service chain supplier over the entire system span. An economical on-line algorithm is proposed, that consists of two main components: 1) A regularization-primarily based approach from Online Learning literature to convert the offline optimal deployment drawback into a sequence of 1-shot regularized issues, each to be efficiently solved in just the once slot and 2) An online dependent rounding scheme to derive feasible integer solutions from the optimal fractional solutions of the one-shot problems, and to ensure a good competitive ratio of the online algorithm over the entire time span. We tend to verify our on-line algorithm with solid theoretical analysis and trace-driven simulations under realistic settings.