Optimization of SLA aware live migration of multiple virtual machines using Lagrange multiplier

Abstract

Cloud computing has attracted customers who want computing as a utility for reducing the overall business cost. For the success of the cloud, it is desirable to have a high degree of trust between the cloud service providers and customers. Service level agreements (SLAs) are major tools to build this trust. Cloud service provider often offers infrastructure-as-a-service, predominantly in the form of virtual machines (VM). These VMs allow customers to run their services in an isolated environment.While provisioning VMs on the cloud, it is often required to migrate multiple virtual machines from one server to another for load balancing, maintenance, or other reasons. This requires allocation of transfer rate to these VMs, which is usually done such that either the total migration time and/or total downtime is minimized without considering the penalty imposed for the service downtime during migration. In this paper, we propose a model for allocation of transfer rate to multiple migrating VMs using Lagrange multiplier such that the SLA violation and consequently the provider’s penalty is minimized during the migration process. To evaluate the model, extensive simulation has been done for various test cases. The results show that the penalty the provider has to pay for the violation of SLA during migration can be reduced significantly with the proposed approach.