Design Optimization for Cloud Application Migration and Remigration

摘要

With the increasing popularity of cloud computing, more and more legacy applications and systems are migrated to clouds, for clouds offer high scalability, stability and low maintenance nature. Meanwhile, since applications are distributed in clouds, reliability and response time of applications could not be guaranteed. To solve the problem, an optimization framework, FLCloud, based on PageRank algorithm is proposed for application migration and remigration. In the algorithm, application structure information, failure rate and statistical response time are considered for the selection of top k components, namely, Web services of applications. Fault tolerance strategy and location modification for these components can improve the whole application reliability and reduce response time of these critical components to enhance user experiences. The experimental results show that the increase of reliability of several significant components according to these QoS properties can improve the dependability of applications in clouds.