K-means algorithm is one of the important algorithms when it comes to clustering. Clustering is a term where all similar kind of data is clubbed together in a cluster. In this paper we have tried to modify the algorithm for its scalability. As this algorithm is mainly used for analyzing big data, when the data is too big it takes hours to run the program and sometimes system can hang. So we will try to reduce its execution time. It will help to get output faster when the data set is large. © IAEME Publication.