Map reduce programming model for parallel K-mediod algorithm on hadoop cluster

This paper presents result analysis of K-Mediod algorithm, implemented on Hadoop Cluster by using Map-Reduce concept. Map-Reduce are programming models which authorize the managing of huge datasets in parallel, on a large number of devices. It is especially well suited to constant or moderate changing set of data since the implementation point of a position is usually high. MapReduce is supposed to be framework of "big data". The MapReduce model authorizes for systematic and instant organizing of large scale data with a cluster of evaluate nodes. One of the primary affect in Hadoop is how to minimize the completion length (i.e., make span) of a set of MapReduce duty. For various applications like word count, grep, terasort and parallel K-Mediod Clustering Algorithm, it has been observed that as the number of node increases, execution time decreases. In this paper we verified Map Reduce applications and found as the amount of nodes increases the completion time decreases.