Design and implementation of performance monitor and evaluation middleware in virtual clouds

Abstract

Cloud computing is an emerging commercial infrastructure paradigm that promises to eliminate the need for maintaining expensive computing facilities. Consequently, more and more cloud systems are designed and deployed to solve large-scale problems in the area of business and science engineering. However, performance monitor and evaluation in cloud environments still remains an open problem, and few efforts are taken to address this issue. In this paper, we present the design and implementation of a middleware, which is aiming to provide users and researchers an easy-to-use toolkit to evaluate or monitor their cloud system's runtime performance. It also can be used as a cloud test-bed for performance analysis when new resource management policies or scheduling algorithms are proposed. Extensive experiments are conducted to investigate the effectiveness of the proposed middleware, and the results indicate that its configurable feature is very useful when users are conducting performance comparing under different conditions.