Aggressive Migration: An Effective Scheduling Policy

Cloud computing is a terminology which is achieved by Distributed and Parallel computing. It offers end consumers an a pay as go model and provides services to remote users. To provide easy customer access, many cloud computing companies integrate frameworks for parallel data processing in their product portfolio. Parallel processes require communication and synchronization among various processes. As a result the CPU utilization of parallel applications of certain nature often decreases as parallelism grows. It is challenging for a data center to achieve a certain level of utilization of its nodes while maintaining the level of responsiveness of parallel jobs. Normally, responsiveness is on the top priority for existing scheduling algorithms. In this paper, we review and classify scheduling algorithms in literature.