Very large databases are required to store massive amounts of data that are continuously inserted and queried. Analyzing huge data sets and extracting valuable pattern in many applications are interesting for researchers. We can identify two main groups of techniques for huge data bases mining. One group refers to streaming data and applies mining techniques whereas second group attempts to solve this problem directly with efficient algorithms. Recently many researchers have focused on data stream as an efficient strategy against huge data base mining instead of mining on entire data base. The main problem in data stream mining means evolving data is more difficult to detect in this techniques therefore unsupervised methods should be applied. However, clustering techniques can lead us to discover hidden information. In this survey, we try to clarify: first, the different problem definitions related to data stream clustering in general; second, the specific difficulties encountered in this field of research; third, the varying assumptions, heuristics, and intuitions forming the basis of different approaches; and how several prominent solutions tackle different problems. Index Terms- Data Stream, Clustering, K-Means, Concept drift