Optimal Service Function Chain Placement Modeling for Minimizing Setup and Operation Cost

Abstract

A service function chaining (SFC) problem is to find out a set of nodes and paths in substrate networks for the placement of virtual network functions (VNFs) so that end-to-end bandwidth and latency requirements are satisfied. We define setup and operation cost, and formulate the SFC placement problem as an Integer Linear Programming (ILP) model with the, objective to minimize the overall costs of setup and operation, while satisfying the bandwidth and CPU constraints of each node. We evaluate performance with different objectives of SFC such as latency minimization and maintenance of the order of VNFs in the SFC.