Performance modeling approach of grid service using Queueing Petri Nets

Performance models are used increasingly throughout the phases of the development and running of grid systems. However, as grid systems grow in size and complexity, building models that accurately capture the different aspects of their behavior becomes a more and more challenging task. As a matter of fact, it is more practical to study grid service performance than to study the whole grid. We present a hierarchical structure of grid service and the concept of grid service execution graph. Then we study how Queueing Petri Net models can be exploited as a powerful performance prediction tool for grid service. A detailed performance model is built in a step-by-step fashion and then used to tackle performance evaluation problem of grid services. In addition to CPU and I/O capacity, it was demonstrated how some complex aspects of grid components behavior, such as service dependency style, reliability impact, and recovery ability, can be modeled. We show that this solution can be helpful to design, analyze and develop grid system more efficiently.