Graph Modeled Oppositional Whale Optimization for Efficient Cloud Service Provision Using Virtual Network Mapping.

Cloud provides preferred services to the users based on their requirements over the internet. In a cloud server, virtual network mapping (VNM) is employed to build a network on demand by deploying virtual machines in a substrate network based on the user request and offers efficient service provision. In case of service provisioning, sufficient computing resource is an essential one to deliver services for satisfying the customers' requirements. The resource optimal service provisioning is a still demanding issue in cloud computing. To develop, Oppositional Learned Metaheuristic Whale Optimization-based Graphical Virtual Network Mapping (OLMWO-GVNM) technique for improving the mapping efficiency of virtual network request to the resource-efficient physical node with minimum computation time. Initially, the number of virtual network requests arrived at the cloud server from different locations. After collecting the requests, the cloud server uses the graph theory for performing both virtual node mapping and link mapping with optimal resource utilization. The node mapping is carried out using the Oppositional Learned Metaheuristic Whale Optimization technique with multiple resources such as CPU, memory, bandwidth, and energy to find the optimal physical node for mapping the virtual network requests. After that, the link mapping is carried out by mapping the virtual link on the substrate path. In this way, the OLMWO-GVNM technique efficiently mapping the input virtual network request to the resource optimum physical node with minimum time. Experimental evaluation of the OLMWO-GVNM technique and existing techniques are carried out with different factors such as mapping efficiency, computation time, request acceptance ratio, and memory consumption concerning several requests. The discussed results prove that the proposed OLMWO-GVNM technique improves the virtual network mapping efficiency and acceptance ratio with minimum time as well as memory consumption compared to state-of-the-art methods. [ABSTRACT FROM AUTHOR]