Automatic detection of multi-level deadlocks in distributed transaction management systems
ABSTRACT A model of asynchronous transaction management has been proposed in this study. This model demonstrates a procedure for elimination of delays caused by the occurrence of distributed deadlocks. The possibilities of occurrence of deadlocks are eliminated by using multiple asynchronous operations. By using the proposed model of activity, many conventional delays associated with transaction processing get eliminated prior to the occurrence of a wait state.
- SourceAvailable from: Michael Richard Hannaford
Conference Proceeding: Optimization of Detected Deadlock Views of Distributed Database.[show abstract] [hide abstract]
ABSTRACT: Deadlock is one of the most serious problems in multitasking concurrent programming systems. The deadlock problem becomes further complicated when the underlying system is distributed and when tasks have timing constraints. Deadlock detection and optimization is very difficult in a distributed database system because no controller has complete and current information about the system and data dependencies. The deadlock problem is intrinsic to a distributed database system which employs locking as its concurrency control algorithm. In this paper, an optimization technique for the detected deadlock is presented which minimizes the abortion of the selected victim transactions. The optimization technique is concerned with the detection of the transactions which are the basis for the most of the deadlock cycles (either local or global) in the system. The presented technique aborts the transaction's requests which are really to blame for the formation of many deadlock cycles. Also the presented deadlock detection algorithm does not detect any false deadlock or exclude any really existing deadlocks. In this technique global deadlock is not dependent on the local deadlock system.Proceedings of the International Conference on Data Storage and Data Engineering, DSDE 2010, Bangalore, India, 9-10 February 2010; 01/2010