Deadlock prevention in a distributed database system
P. Krishna Reddy, Subhash Bhalla
Abstract
P. Krishna Reddy, Subhash Bhalla
Abstract
The distributed locking based approaches to concurrency control in a distributed database system, are prone to occurrence of deadlocks. An algorithm for deadlock prevention has been considered in this proposal. In this algorithm, a transaction is executed by forming wait for relations with other conflicting transactions. The technique for generation of this kind of precedence graph for transaction execution is analyzed. This approach is a fully distributed approach. The technique is free from deadlocks, avoids resubmission of transactions, and hence reduces processing delays within the distributed environment.
OpenAlex reports 18 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The distributed locking based approaches to concurrency control in a distributed database system, are prone to occurrence of deadlocks. An algorithm for deadlock prevention has been considered in this proposal. In this algorithm, a transaction is executed by forming wait for relations with other conflicting transactions. The technique for generation of this kind of precedence graph for transaction execution is analyzed. This approach is a fully distributed approach. The technique is free from deadlocks, avoids resubmission of transactions, and hence reduces processing delays within the distributed environment.
Key concepts: Computer science, Concurrency control, Distributed transaction, Serializability, Deadlock, Distributed computing, Database transaction, Deadlock prevention algorithms