Throughput maximization and resource optimization in NFV-enabled networks

Network function virtualization (NFV) has been emerging as a new paradigm to enable elastic and inexpensive network services in modern computer networks, through deploying flexible virtualized network functions (VNFs) running in virtual computing platforms. Different VNFs can be chained together to form different service chains, to meet various user data routing demands for different network services. In this paper we consider provisioning network services in an NFV-enabled network that consists of data centers for implementing VNF instances of service chains and switches. We study the throughput maximization problem with the aim to admit as many user requests as possible while minimizing the implementation cost of the requests, assuming that limited numbers of instances of each service chain have been stored in data centers. We first propose an optimal algorithm for the problem if all requests have identical packet rates; otherwise, we devise two approximation algorithms with probable approximation ratios, depending on whether the packet traffic of each request is splittable. We finally conduct experiments to evaluate the performance of the proposed algorithms by simulations. Experimental results show that the proposed algorithms achieve at least 15% more throughput than that of a greedy algorithm.