A comparison of most recent MapReduce joins algorithms

In this interesting line of research, an attempt has been to overview different parallel processing platforms that implement MapReduce jobs. This survey provides a wide-ranging analysis of work and publications related to MapReduce framework to data, and it also can be used as a basis for further research and examination. The scope of this survey is focused on pre-processing, pre-filtering, partitioning, replication, load balancing, performance, memory space, communication cost, and query processing and optimization aspects in the light of big data analysis in MapReduce. Moreover, a set of efficient optimized and improved approaches in the context of analytical query processing and optimizing using MapReduce. It provides an added value to current research published yearly by introducing a comprehensive classification of recently presented papers in the era of join types using MapReduce. From data-centric perspective, the main topic of this approach is intended to highlight the importance of traditional problems of data management and analysis in the regard of efficient big data processing and analysis approaches.