Distinction of Discrete Transformations Applied to Hadoop's MapReduce
Bhavini Kanoongo, Puja Jagani, Chetashri Bhadane · International Journal of Computer Applications · 2014
Hadoop MapReduce is an effective data processing platform for both commercial as well as academic applications.It intends the simplification of vast quantities of data as well as ease of processing in parallel on enormous clusters of hardware in a fault-tolerant and dependable approach.There are many modifications possible in the MapReduce to increase the performance along with increasing the simplicity of job tuning.Three of the adaptive run-time techniques namely, HIPI (Hadoop Image Processing Interface), HOG (Distributed Hadoop MapReduce on Grid) and using SAM (Situation Aware Mappers) are described and compared in the following paper.There is a rapid increase in the amount of images uploaded on the internet; however the applications which utilize this data are severely inadequate.Large and computational distributed processing can be done by employing HIPI.Another Hadoop transformation that we study is the HOG which provides a complimentary, adaptable and dynamic MapReduce environment on the resources of the grid, reforms Hadoop's fault tolerance for wide area data analysis.All the modifications to the Hadoop framework are transparent to the existing Hadoop MapReduce applications.