Abstract
In Data analysis process clustering is one of the standard methods, which is used in many area
such as pattern recognition and image segmentation and statistics, bioinformatics etc. Clustering have several
algorithms in that most well know simplest algorithm is K-means, because of its simplicity, empirical success
and efficiency. This present reality applications produce tremendous volumes of data, subsequently, how to
productively deal with these data in a significant mining task has been a difficult and huge issue. Likewise
Message passing Interface (MPI) as a Programming model for increases the scalability, performance and
execution speed. Enthused by this, MPI with parallel K- means clustering algorithm is implemented in this
paper. The efficiency of the K means clustering algorithm is improved in parallel environment by implementing
in MPI methodology. In this paper performance of the clustering the data using K- means is compared between
sequentially run of K- means algorithm and parallel run of K-means algorithm in Message passing interface
architecture in terms of overhead cost and execution

Key words: K-Means; Message passing Interface; Sequential; parallel computing.