Abstract: Clustering is the unsupervised classification of patterns (data items) into groups (clusters).Clustering in data mining is very useful to discover distribution patterns in the underlying data. Today, the term "a large dataset" refers to hundreds of terabytes or even petabytes of data. This type of datasets are too difficult to for a clusters. It is typical of scientific investigations to have two phases: the data generation phase, and the data analysis phase. The data generation phase is usually the result of running a large simulation or the collection of data from experiments. It is desirable to design an ant colony optimization algorithm (ACO)[6][4][5] that is not required to solve any hard sub problem but can give nearly optimal solutions for data clustering. The proposed method can obtain optimal solutions quicker via differently favorable strategy. In this paper, we present a new data clustering method for data mining in large databases. Our simulation results show that the proposed Clustering huge datasets(CLHDS) method performs better than the Fast SOM combines K-means approach (FSOM+K-means) and Genetic K-Means Algorithm (GKA,K-Medoids algorithm.