Concurrency Control: Traditional Approaches
Concurrency control; Page locking; Scheduling; Transaction execution; Two-phase locking
The core requirement on a transactional server is to provide the ACID properties for transactions, which requires that the server includes a concurrency control component as well as a recovery component. Concurrency control essentially guarantees the isolation properties of transactions, by giving each transaction the impression that it operates alone on the underlying database and, more technically, by providing serializable executions. To achieve serializability, a number of algorithms have been proposed. Traditional approaches focus on the read-write model of transactions and devise numerous ways for correctly scheduling read-write transactions. Most practical solutions employ a variant of the two-phase locking protocol.
- 1.Bernstein PA, Hadzilacos V, Goodman N. Concurrency control and recovery in database systems. Reading: Addison-Wesley; 1987.Google Scholar
- 3.Claybrook B. OLTP – online transaction processing systems. New York: Wiley; 1992.Google Scholar
- 6.Weikum G, Vossen G. Transactional information systems – theory, algorithms, and the practice of concurrency control and recovery. San Francisco: Morgan Kaufmann; 2002.Google Scholar