Measuring the capacity of a streaming media server in a Utility Data Center environment

Abstract

Abstract In order to design a "utility-aware" streaming media service which automatically requests the necessary resources from Utility Data Center infrastructure, several classic performance questions should be answered: how to measure the basic capacity of a streaming media server? what is the set of basic benchmarks exposing the performance limits and main bottlenecks of a media server? In this paper, we propose a set of benchmarks for measuring the basic capacities of streaming media systems for different expected workloads, and demonstrate the results using an experimental testbed.