Study on emerging implementations of MapReduce

MapReduce is a programming model specifically developed for the management and processing of “Big Data” - extremely large amounts of data that expects high level of analyzing capabilities. With every passing day volumes of data is generated and collected from multiple data resources across the planet. This data must be analyzed in the sense of volume or speed of data moving to and from the data management systems. MapReduce efficiently execute programs on large clusters by utilizing the concept of parallelism. Till now Google´s MapReduce framework has been considered as the most successful implementation for Big Data. A number of implementations of MapReduce programming model have been proposed. This paper discusses various emerging implementations of MapReduce model. An emphasis is also given on the leading and lacking strength of these implementations.