An Efficient Representation Using Harmony Search for Solving the Virtual Machine Consolidation

Abstract

A data center with a large number of servers, large storage, and many network devices requires power for cooling to reduce heat generation, air conditioning, and emergency power generation facilities, in addition to power for operation internally consumed by infrastructure equipment. The power consumed by data centers worldwide makes up a large proportion. Although the size of data centers is expected to increase, we are already faced with power problems because stability is prioritized over efficiency when operating data centers in order to meet the Service Level Agreement (SLA) conditions. Most data centers are in a virtualization environment, and virtual machine consolidation using physical machine (PM) transitions to the idle mode through virtual machine (VM) migration has been suggested as one of the most effective ways to reduce the amount of power usage in a data center. This study takes into account the characteristics of virtualization environments and presents an algorithm that effectively solves VM consolidation (VMC) through operator design using a grouping representation method and a meta-heuristic method known as harmony search.

Nature-Inspired Optimization for Virtual Machine Allocation in Cloud Computing: Current Methods and Future Directions

“…However, it ignores scalability and resource utilization. www.ijacsa.thesai.org Kim, et al [57] have proposed a new method using a grouping agent approach and a meta-heuristic harmonic search algorithm for effectively solving the virtual machine allocation problem. In this algorithm, to solve VMC, the population number should be increased in proportion to increasing the size of the problem to certify search variety.…”