With relations growing larger, distributed, and queries becoming more complex, parallel query processing is an increasingly attractive option for improving the performance of database systems. Distributed and parallel query processing has been widely used in data intensive applications where data of relevance to users are stored at multiple locations. In this paper, we propose a three-tier middleware system for optimizing and processing of distributed queries in parallel on Cluster Grid architecture. The main contribution of this paper is providing transparent and integrated access to distributed heterogeneous data resources, getting performance improvements of implicit parallelism by extending technologies from parallel databases. We also proposed the dynamic programming algorithm for query optimization and site selection algorithm for resource balancing. An example query for employee databases is used throughout the paper to show the benefits of the system.