A comparative study of concurrency control algorithms for distributed databases

The declining cost of computer hardware and the increasing data processing needs of geographically dispersed organizations have led to substantial interest in distributed data management. These characteristics have led to reconsider the design of centralized databases. Distributed databases have appeared as a result of those considerations. A number of advantages result from having duplicate copies of data in a distributed databases. Some of these advantages are: increased data accesibility, more responsive data access, higher reliability, and load sharing. These and other benefits must be balanced against the additional cost and complexity introduced in doing so. This thesis considers the problem of concurrency control of multiple copy databases. Several synchronization techniques are mentioned and a few algorithms for concurrency control are evaluated and compared