Designing a High Performance Integrated Strategy for Secured Distributed Database Systems

摘要

The increasing complexity and rapid growth of the real world applications resulted in increasing interest in discovering useful methods in Distributed Database systems (DDBS) to provide users with tools to effectively and efficiently process and control information.rnData fragmentation techniques have been used in partitioning database and improving the DDBS performance throughout by reducing the number of data packets (fragments) to be accessed. A precise data allocation decision minimizes the cost of data transfer and reduces the number of messages required during applications processing. Clustering mechanisms provide a structure for organizing large number of sites into a set of useful clusters for efficient transaction processing operations.rnThis chapter presents an integrated strategy for distributed relational database design that efficiently and effectively achieves the objectives of data fragmentation, data allocation and clustering mechanisms. It has been designed to split the database relations into pair-wise disjoint fragments and find out precisely; based on a high performance clustering technique and transaction processing cost functions, whether each fragment has to be allocated, where the allocation benefit outweighs the cost, or removed from each cluster and its sites respectively. This can reduce the amount of redundant data to be accessed and transferred among different sites, significantly improve transaction performance and overall system response time, and result in better decision support.rnExperimental studies on real database applications at different networks connectivity are performed and the results demonstrate that the proposed strategy leads to precise solutions for the problems of data fragmentation and allocation, and better set of clusters in comparison with other clustering methods. This serves to illustrate the efficiency of the strategy.