Clustering is division of data into groups of similar objects. Each group called cluster, consists of objects that are similar amongst them and dissimilar compared to objects of other groups. A comparative study of clustering algorithms across three different datasets is performed. The algorithms under investigation are partitioning based i.e. k-means, farthest first, expectation maximization and non-partitioning based i.e. density based, hierarchical based and cobweb. All these algorithms are compared according to the factors size of the dataset, number of clusters and time taken to form clusters.