PROJECT TITLE :

Network Function Placement for NFV Chaining in Packet/Optical Datacenters

ABSTRACT:

In an operator's datacenter, optical technologies can be employed to perform network perform (NF) chaining for larger aggregated flows in parallel with the traditional packet-based mostly fine-grained traffic steering schemes. When network function virtualization (NFV) is enabled, virtualized NFs (vNF) can be placed when and where needed. During this study, we establish the possibility of minimizing the expensive optical/electronic/optical (O/E/O) conversions for NFV chaining in packet/optical datacenters, that is introduced by the on-demand placement of vNFs. When the vNFs of the identical NF chain are properly grouped into fewer pods, traffic flows can avoid unnecessary traversals in the optical domain. We have a tendency to formulate the problem of optimal vNF placement in binary integer programming (BIP), and propose another economical heuristic algorithm to solve this drawback. Evaluation results show that our algorithm will achieve near-optimal O/E/O conversions love BIP. We have a tendency to additionally demonstrate the effectiveness of our algorithm under various eventualities, with comparison to a simple initial-work algorithm.