Enhancing Dynamic Capacity Scheduler for Data Intensive Jobs

Management of Big Data is a Challenging issue. The MapReduce environment is the widely used key solution for data intensive jobs. We will analyze map reduce pipelining and along with processing of Map phase and Reduce phase. Core schedulers FIFO, Fair and Capacity Schedulers have been discussed. The Scheduler assigns MapReduce task to the resources and there is a challenge to the scheduler to schedule the task in a way that it is scalable. Existing work shows the performance of the Hadoop depends upon input data and configuration of the cluster. In this paper, we have analyzed the execution time for data intensive jobs with increasing volume of the data set. We have also compared the execution time of the task with existing scheduler and our proposed method for the scheduler