DYNAMICAL PROCESSING TECHNIQUE FOR DATA FRAGMENTATION IN DISTRIBUTED DATABASE SYSTEMS: DESIGN AND TOOL SUPPORT

Data fragmentation is one of the primary techniques that has been used in partitioning and developing distributed database systems (DDBs). Database partitioning technique improves the performance of the system throughput and reduces the amount of irrelevant data to be accessed and transferred among different sites in DDBs. We propose a technique that splits the database relations into pair-wise disjoint fragments, which would be allocated to the DDBs sites during the process of allocation. This technique reduces the data transferred between the sites during the applications execution, minimizes the communication cost needed for processing applications, handles the database queries and meets their future needs, guarantees the ability of processing a given portion of a given transaction in all sites, and improves the applications speed and efficiency by minimizing the number of fragments to be accessed. We have also developed a fragmentation tool that is interactive, full functional and easy to use, based on our fragmentation life cycle algorithms to create the database segments and generate the minimum number of disjoint fragments. A case study and some results in tabular and graphical means will be presented to demonstrate the efficiency of the method.