Many modern applications generate data streams which are time-ordered, continuous, and unbounded in nature and are generated at a rapid rate. Processing these streams requires more memory, the results need to be generated faster and the incoming data streams can only be viewed once. Traditional algorithms cannot be used for mining streaming data. Therefore, in this project optimizing processing of multiple queries (OPMQ) framework is being implemented for simultaneously executing multiple queries according to the inherent commonalities within. The input data set is converted into different clusters so that the accessing time of the data is reduced thereby improving the performance of the system. The final executable clusters are generated using the k-means algorithm which uses Euclidean distance as the heuristic to find the nearest cluster centroid.