Big data analytics as well as data mining are plays vital role in extracting the hidden statistics. Customary advances for investigation & extraction of hidden information from data may not exertion efficiently for big data since of its complex, very elevated volume nature. Data clustering is single of the data mining technique which exacts the useful data from the data by grouping data into clusters. In Big data as the data is complex and of very large volume, individual clustering techniques may not consider all the samples it may leads to inaccurate results. To overcome this inaccuracy this proposed method is the combination of dynamic k-means and hierarchical clustering algorithms. This proposed method can be called as hybrid method. Being hybrid method will overcome few drawbacks like static k value .In this paper proposed method is compared with existing algorithms by using some clustering metrics.