A PERFORMANCE STUDY OF HARDWARE IMPACT ON FULL VIRTUALIZATION FOR SERVER CONSOLIDATION IN CLOUD ENVIRONMENT

Underutilization of hardware resources has always b een a problem in single workload driven traditional OS environment. To improve resource utilization, virtu alization of multiple VMs and workloads onto the sa me host with the aid of Hypervisor has been the recent trend. Its use cases such as server consolidation, live migration, performance isolation and on-demand serv er provisioning make it as a heart part of enterpri se application cloud. Cloud is an on-demand, service p rovisioning technology, where performance plays a vital role for user acceptance. There are numerous virtualization technologies are available from full virtualization to paravirtualization, each has its s rength and weakness. As performance study is an o ngoing pursuit, hardware and software development getting matured day by day, it is desirable to do this sort of performance study in regular interval that often sh eds new light on aspects of a work not fully explor ed in the previous publication. Hence, this paper focus p erformance behaviours of various full virtualizatio n models such as hosted (VirtualBox) and bare metal ( KVM) virtualization using variety of benchmarks fro m micro, macro and application level in the cloud en vironment. We compare both virtualization solutions with a base system in terms of application performa nce, resource consumption, low-level system metrics like context switch, process creation, interprocess communication latency and virtualization-specific metrics like virtualization layer consumption. Expe rimental results yield that VirtualBox outperforms KVM in CPU and thread level parallelism and KVM outperf orms in all other cases. Both are very reluctantly accepted for disk usages comparing with native syst em.