Benchmarking Grid Applications for Performance and Scalability Predictions

Application benchmarks can play a key role in analyzing and predicting the performance and scalability of Grid applications, serve as an evaluation of the fitness of a collection of Grid resources for running a specific application or class of applications [25], and help in implementing performance-aware resource allocation policies of real time job schedulers. However, application benchmarks have been largely ignored due to diversified types of applications, multi-constrained executions, dynamic Grid behavior and heavy computational costs. To remedy these, we present the GrapBench (Grid Application Benchmarks) system. GrapBench computes the Grap benchmarks for Grid applications which are flexible regarding variations in problem-size of the application and machine-size of the Grid-site. GrapBench dynamically controls the number of benchmarking experiments for individual applications, and manages the execution of these experiments on different Grid-sites in an easy and flexible way. We also present results from the prototype implementation of our proposed system to show the effectiveness of our approach