Improving Heterogeneous Workload Performance in Server Virtualization through User Behavior Schedules

One of the most attractive topics today is servervirtualization. Virtualization is a method of running multipleindependent virtual operating systems on a single physicalcomputer. It is a way of maximizing physical resources tomaximize the investment in hardware. This concept can help toconsolidate servers, improve hardware utilization and reducethe consumption of power and physical space in the datacenter. However, in an environment where workloads sharethe underlying platforms, the problem of configurationcontrols on sharing dynamic resource allocation becomes achallenge. In this paper, we present the improved performanceon heterogeneous workload in server virtualization using userbehavior schedules. We show a new concept for managingworkloads by involving user behavior in each type of serviceworkload. The result shows that user behaviors are different ineach type of service workload and time. This preliminary studymay be an approach to improve management of data centersrunning heterogeneous workloads using server virtualization.