Building a Performance Model of Streaming Media Applications in Utility Data Center Environment

Abstract Utility Data Center (UDC) provides a exible, coste ective infrastructure to support the hosting of applications for Internet services. In order to enable the design of a \utilityaware" streaming media service which automatically requests the necessary resources from UDC infrastructure, we introduce a set of benchmarks for measuring the basic capacities of streaming media systems. The benchmarks allow one to derive the scaling rules of server capacity for delivering media les which are: i) encoded at di erent bit rates, ii) streamed from memory vs disk. Using an experimental testbed, we show that these scaling rules are nontrivial. In this paper, we develop a workload-aware, media server performance model which is based on a cost function derived from the set of basic benchmark measurements. We validate this performance model by comparing the predicted and measured media server capacities for a set of synthetic workloads.

• Type: Conference
• Year: 2003
• Where: CCGRID
• Authors: Ludmila Cherkasova, Loren Staley
• Keywords: CCGRID 2003, Cluster Computing, Media Server, Scaling Rules, Streaming Media