In this paper, two algorithms were introduced to describe two algorithms to describe and compare the applying of the proposed technique in the two types of the distributed database system. The First Proposed Algorithm is Homogeneous Distributed Clustering for Classification (HOMDC4C), which aim to learn a classification model from unlabeled datasets distributed homogenously over the network, this is done by building a local clustering model on the datasets distributed over three sites in the network and then build a local classification model based on labeled data that produce from clustering model. In the one computer considered as a control computer, we build a global classification model and then use this model in the future predictive. The Second Proposed Algorithm in Heterogeneous Distributed Clustering for Classification (HETDC4C) aims to build a classification model over unlabeled datasets distributed heterogeneously over sites of the network, the datasets in this algorithm collected in one central computer and then build the clustering model and then classification model. The objective of this work is to use the unlabeled data to introduce a set of labeled data that are useful for build a classification model that can predict any unlabeled instance based on that classification model. This was done by using the Clustering for Classification technique. Then presented this technique in distributed database environment to reduce the execution time and storage space that is required.