Performance Prediction for SaaS Deployment Optimization Based on Colored Petri Nets

The deployment of SaaS (Software-as-a-Service) application in clouds influents its performance, and thus it is necessary to estimate the performance of SaaS application when optimizing SaaS deployment. In this paper, a colored petri nets (CPN) based SaaS application performance prediction approach for deployment optimization is proposed. Specifically, a CPN modeling approach for SaaS application with replaceable deployment is proposed; a performance prediction technique based on simulation of a CPN model is explored. Moreover, experiments in e-commerce domain are conducted, and the results validate the effectiveness of the approach.