Performance Improvement Techniques for MapReduce - A Survey

Enterprises these days acquire huge volumes of data from totally different sources and leverage this information by means that of data analysis to support effective decision-making and supply new practicality and services. The key demand of data analytics is scalability, merely thanks to the vast volume of information that requires to be extracted, processed, and analyzed in a very timely fashion. Arguably the foremost widespread framework for modern large-scale data analytics is MapReduce, in the main thanks to its salient features that embrace scalability, fault-tolerance, simple programming, and flexibility. However, despite its deserves, MapReduce has evident performance limitations in miscellaneous analytical tasks, and this has given rise to a major body of work that aim at up its potency, whereas maintaining its fascinating properties. This survey aims to review the state of the art in improving the performance of MapReduce.