Analysis of MapReduce Model on Big Data Processing within Cloud Computing

Nowadays cloud computing is becoming a trend on big data processing. Google created MapReduce model to simplify the complex computation of big data processing by configuring and splitting the data into key/values pair to be processed in parallel, usually within a network of computers, then merge the results. However, MapReduce model has its limitations. Researchers have been trying to improve the model resulting in some newer models, such as Mantri, Camdoop, Sudo, and Nectar model. Each model exploits the different characteristics of MapReduce model to create improvements in different way and cases. Challenges and improvements still remain within these enhanced models, which open new possibilities on area of research