சுருக்கம்

Enhancing Map-Reduce Mechanism for Big Data with Density-Based Clustering

Vinod S. Bawane, Deepti P. Theng

Map-Reduce is software framework that allows certain type of parallelizable or distributable problems involving bulky data sets to be solve using computing clusters. This paper presents a hybrid Map-Reduce framework that gathers computations resources from different clusters and runs Map-Reduce jobs across them. The mechanism is realized using DBSCAN clustering algorithms among Map-Reduce, parallel processing framework over clusters. However, the instant accomplishment of algorithms undergoes from efficiency problem for higher execution time as well as inadequate memory. This paper presents an efficient DBSCAN clustering method for mining large datasets with apache Hadoop and Map-Reduce that will reduce time of accessible algorithms and dataset from the dissimilar location will work simultaneously from single node and find the appropriate outcome in distributed environment