Abstract: MapReduce is one of the programming model for processing large amount of data in cloud where resource allocation is one of the research areas since it is responsible for improving the performance of Hadoop.Many resource allocation strategies are discussed that aims to improve resource utilization in Hadoop by focusing on the speculative execution and maximizing the data locality by meeting the deadlines of the job. The implementation is done in Hadoop slot based MapReduce system ex: Hadoop MVR1.

Keywords: MapReduce, Data locality, Speculative execution, Hadoop.