An implementation of virtualization cluster: Extending Beowulf cluster using virtualization cluster management and image storage

Computer clusters have become a main stream of high performance computing platforms. Multilevel computing engines are spanned across acceleration boards, multicore processors, multiprocessor boards, and computer clusters. Multicore technology have played major role in increasing density of parallel computing elements. To harvest these high performance systems, cluster operating environment has pressed on additional abstraction using virtualization technology. Specific problem solving environments are isolated at the operating system level where real executions are performed in virtualization domain. Virtualization technology helps not only increasing utilization of computing resources but also reducing configuration workload, administrative cost, application porting and energy saving. Multiple commodity operating systems are allowed to share conventional hardware in a safe environment. In this work, we investigate an implementation of virtualized computing cluster. Performance evaluation results of the virtualized cluster based on HPL is shown where the maximum of 20% performance degradation from virtualization overhead is observed. We describe straightforward approach to deploy virtualization onto the cluster. The management of virtualization images is discussed. The process of considering computational cluster resources in virtualization environment is described. The goal of this work is to show the virtualized resources running with production of virtual cluster environment.