Abstraction based K-tier partitioning scheme for parallel job scheduling in multiple cloud centers

In large scale super computing environment, parallel machines have traditionally used space sharing strategies to accommodate multiple jobs at the same time by dedicating the nodes to a single job until it completes. Existing scheduling schemes make use of priority based method for consolidating parallel workloads in cloud. This results in starvation of larger jobs, reduced throughput and underutilization of resources. In this paper, Abstraction based K-tier Partitioning scheme is proposed for efficient scheduling of jobs in k- cloud data center with multiple computing capacities by solving large-scale static scheduling problem using abstraction refinement.