SERCON-BASED TIMESTAMPED VIRTUAL MACHINE MIGRATION SCHEME FOR CLOUD

With the advent of cloud computing, the need for deploying multiple virtual machines (VMs) on multiple hosts to address the ever-increasing user demands for services has raised concerns regarding energy consumption. Considerable energy is consumed while keeping the data centers with a large number of servers active. However, in data centers, there are cases where these servers may not get utilized efficiently. There can be servers that consume sufficient energy while running resources for a small task (demanding fewer resources), but there can also be servers that receive user requests so frequently that resources may be exhausted, and the server becomes unable to fulfill requests. In such a scenario, there is an urgent need to conserve energy and resources, which is addressed by performing server consolidation. Server consolidation aims to reduce the total number of active servers in the cloud such that performance does not get compromised as well as energy is conserved in an attempt to make each server run to its maximum. This is done by reducing the number of active servers in a data center by transferring the workload of one or more VM(s) from one server to another, referred to as VM Migration (VMM). During VMM, time is supposed as a major constraint for effective and user-transparent migration. Thus, this paper proposes a novel VM migration strategy considering time sensitivity as a primary constraint. The aim of the proposed Time Sensitive Virtual Machine Migration (TS-VMM) is to reduce the number of migrations to a minimum with effective cost optimization and maximum server utilization.

Automated summary

With the rise of cloud computing, there is a need to deploy multiple virtual machines (VMs) on multiple hosts. However, the energy consumption of data centers with a large number of servers remains a concern. Server consolidation aims to reduce the number of active servers in a data center by transferring the workload of VMs from one server to another. This paper proposes a Time Sensitive Virtual Machine Migration (TS-VMM) strategy to minimize migrations while optimizing cost and maximizing server utilization.