Optimization of Infrastructure as a Service Model in Cloud Computing

Abstract

The proposed work focuses on Cloud Interoperability and SLA negotiation, among these issues. A Virtual Machine is a compute resource hosted on servers to manage user requests. In the community cloud, when the provider cannot satisfy the user request, that VM is migrated to another provider in the group. VM migration is needed to reduce energy consumption, fault management, low level system maintenance, and load balancing. The cloud services are available to the users depending on the SLA between the cloud providers and the users. The SLA violation detection mechanism is advantageous for enhancing trust, avoiding penalty charges, and increasing the cloud providers profit. In this research, two significant contributions are presented to selecting a provider for VM migration in community cloud computing. newlineFirst, broker policy is a critical decision factor for cloud computing. During this process, the broker allocates the cloudlets to the different data centres. In this paper, we have considered four strategies for allocating cloudlets to the data centre. The broker allocation policies are round robin, Bee colony optimization, genetic learning, and particle swarm algorithm. Finally, the Enhanced genetic learning based Particle swarm optimization technique is designed and tested for performance. The Enhanced genetic learning based particle swarm optimization technique shows an improvement of 10 percentage over the other soft computing techniques. newlineSecond, virtual machine placement is the task of planning the optimal map of virtual machines to physical machines. With the help of optimal allocation of virtual machines, there is a chance of lowering the power consumption and reducing ineffective use of resources The algorithms objective is to reduce the wastage of resources and reduce power consumption. Resources and power consumption are constrained to the parameter that the total bandwidth required by the application should not be disturbed. The proposed solution based on NSGA III gave better performance than MGGA. newline