Scheduling Strategies in Hadoop: A Survey

AbSTRACT Hadoop-MapReduce is one of the dominant parallel data processing tool designed for large scale data-intensive cloud computing platforms. Hadoop Mapreduce framework requires a proficient mechanism to allocate and schedule computing resources which makes it liable to accomplish the desired preferences for the specific business application. In this paper, an analysis of a few scheduler improvement methods that improves the completion time of the job or the task has been carried out in order to facilitate a judicious choice of the scheduler based on the requirement and application expectations