Reliability-aware DAG scheduling with primary-backup in cloud computing

Abstract

Cloud computing environments facilitate applications by providing virtualised resources that can be provisioned dynamically. The efficient scheduling of application tasks is critical to achieve high performance in cloud computing systems. To schedule precedence graphs in a more realistic cloud computing environments, we introduce a new scheduling model that considers the contention for communication not only processors resource but also network interface, what's more it is more suitable for fault-tolerant scheduling. Another novel idea proposed in this paper is taken primary-backup replication to duplicate all tasks and get an optimal scheduling solution. Our algorithms have a low time complexity, and drastically improve reliability by replication mechanism and reliability estimation. Experimental results fully demonstrate the usefulness of the proposed algorithms, which lead to efficient execution schemes while guaranteeing a prescribed level of fault tolerance.