期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2012
卷号：1
期号：7
页码：225-228
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Predicting performance of a software system with bursts of customer sessions is essential to ensure scalability and availability and manageability. The tools that can predict performance can help in supporting systems in terms of planning, management and system sizing. We propose a new algorithm for predictive performance. That is WAM (Weighted Average Method) which is meant for improving accuracy of predictions. WAM considers bursts of customer requests in order to reflect the influence of bursts in the system. The experimental results revealed that WAM is efficient in predicting performance of selected software system with bursts of customer requests.
