A taxonomy of performance prediction systems for parallel and distributed computing systems
Sena Seneviratne, DC Levy, Rajkumar Buyya · 2015
© 2015 by Nova Science Publishers, Inc. All rights reserved. As Distributed Computing Systems (DCSs) such as Clusters, Parallel Systems, distributed database systems, Peer-to-Peer Desktop Systems, Grids and Clouds are congregations of geographically distributed heterogeneous resources, the efficient scheduling/ utilization of the resources requires the support of sound Performance Prediction Systems (PPS). The performance prediction of DCS resources is helpful for both Resource Management Systems and users to make optimized resource usage decisions. In this chapter we focus on the taxonomy for the Grid PPS architecture. The taxonomy is used to categorize and identify approaches which are followed in the implementation of the existing PPSs of DCSs such as Clusters, Parallel Systems, Peer-to-Peer Desktop Systems and Grids. The taxonomy and the survey results are used to identify approaches and issues that have not been fully explored in research.