FH-ACO: Fuzzy heuristic-based ant colony optimization for joint virtual network function placement and routing

Network function virtualization (NFV) is a new networking paradigm, which replaces the specific-purpose hardware appliances with software virtualization to perform network functions on general-purpose servers. Due to resource limitation at network servers/links, virtual network function (VNF) placement/routing is one of the main challenges in the NFV architecture. The VNF placement/routing problem has been reported to be NP-hard, and consequently, most of studies have focused on designing heuristics or metaheuristics. In this paper, fuzzy heuristic-based ant colony optimization (named FH-ACO) is introduced as a fuzzy knowledge-based version of ACO to efficiently solve the VNF placement/routing problem. Our motivation is to simultaneously gain with fast speed of heuristics and high solution quality of metaheuristics. In the FH-ACO, two multi-criteria fuzzy inference systems are used as heuristic information to guide artificial ants in appropriate server/link selection during the solution construction phase. To solve the problem, a multi-objective function comprising power consumption, latency, reliability, and load balancing, is used. The proposed FH-ACO algorithm is an application-specific protocol, which can be adaptively tuned based on the application requirements. Simulation results on US and Pan-European network topologies demonstrate the superiority of the proposed FH-ACO algorithm against the existing techniques. (C) 2021 Elsevier B.V. All rights reserved.