Virtualization will dramatically change enterprise system deployments and is being driven by innovation across a broad set of platform technologies. All of this attention has created a broad interest in comparing different products. One challenge is that there are no established performance methodologies to measure virtualization performance. Today's existing server benchmarks cannot be easily used as-is by an end user to generate clear and relevant results. This paper presents a workload methodology to help the reader characterize the performance of servers exploiting virtualization technologies to consolidate multiple physical servers. It presents an example benchmark using applications often found in typical virtualization deployments. The existence and exploitation of a standard methodology is a key to accelerating continued improvement of virtualization technologies.