Data Allocation and Transaction Processing in Distributed Database Systems

Distributed database processing requires the transmission of data between computers in a distributed system. Data transmission is becoming an increasingly attractive data-dissemination method for large database systems. In order to effectively utilize distributed query processing, it is necessary to have efficient dynamic allocation algorithms that can balance individual and overall performance and can scale in terms of table and query size, incoming transactions and routing bandwidth. We investigate dynamic transaction processing strategies for locally distributed database systems in which the database is partitioned and distributed among multiple transaction processing systems and the incoming transactions are routed by a common front-end computer system. In view of this fact, we describe new dynamic transaction processing methods which take into account routing history and minimizes the estimated response time of incoming transactions and can be tuned to trade off average and worst-case waiting time.