A Multi-criteria Decision Making Approach for Scaling and Placement of Virtual Network Functions

This paper investigates the joint scaling and placement problem of network services made up virtual functions (VNFs) that can be provided inside a cluster managing multiple points presence (PoPs). Aiming at increasing VNF service satisfaction rates minimizing deployment cost, we use both transport cloud-aware as well multi-attribute decision making (MADM) algorithms for cluster. The original is known to NP-hard hence solved by separating problems solving them individually. experiments are done using dataset containing information deployed digital-twin service. These show considering cloud parameters during perform more efficiently than only based or followed algorithms. One MADM algorithms, Total Order Preference Similarity Ideal Solution (TOPSIS), has shown yield lowest cost highest request compared other investigated Our simulation results indicate in various availability scenarios resources significant potential provide increased when TOPSIS scheme performed.