Paper Details

Technology’s role in society today has a major impact on our overall sense of living and thatis why in the 21st century, it is offered as a subject. The fast and improved speed of computer systems aremaking humans life easier and giving him new opportunity to create an impossible. To improve theprocessing speed of systems different technologies are now used like distributed system, parallelcomputing. The map reduce which is used in parallel computing is one of the popular data model forhigh speed computation in computation technology. The Existing map reduce focuses on scheduling atthe task-level. But unfortunately, the task-level scheduling leads to inefficient job schedules with lowresource utilization and long job execution time.In this concept we divide the tasks into unequal parts called as phases and apply phase-level schedulingto these phases and achieve efficient resource usage. In this paper, we present a Scheduler, a Phase andResource Information-aware Scheduler for MapReduce clusters that performs resource-awarescheduling at the level of task phases. Specifically, we show that for most MapReduce applications, therun-time task resource consumption can vary significantly from phase to phase. Therefore, byconsidering the resource demand at the phase level, it is possible for the scheduler to achieve higherdegrees of parallelism while avoiding resource contention.

Nisha Shinde, Trupti Patil, Prajkta Shinde, Himani Mavchi, “Efficient Job Execution for Map Reduce Using Phase-Level Scheduling Algorithm.”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 5, pp. 665-671, May 2017.