DEV Community

Arunesh Choudhary
Arunesh Choudhary

Posted on

05 - Concurrency Control

In a multi-user environment where numerous transactions may access and edit the same data concurrently, concurrency control is an essential component of database management systems (DBMS) that assures data consistency and integrity. By coordinating the execution of transactions, concurrency control systems avoid conflicts and protect data integrity.

The following are some significant theories and methods in concurrency control:

Transactions: Logical units of work called transactions are made up of one or more database operations. It symbolizes a series of activities that should be carried out atomically, which means that either all of them succeed or none of them have any impact at all.


Isolation Levels: The level of transactional isolation is determined by the use of isolation levels. Different levels of isolation offer various concurrency and consistency guarantees. READ UNCOMMITTED, READ COMMITTED, REPEATABLE READ, and SERIALIZABLE are examples of common isolation levels.

Locking: Locking is a common concurrency control method. To limit access and modification, it entails obtaining and releasing locks on database objects. Locks can be shared (read locks) or exclusive (write locks). By ensuring that only one transaction may alter a data item at once, locking helps to avoid conflicts and preserve consistency.

Multiversion Concurrency Control (MVCC): MVCC is a locking option that is frequently employed in isolation levels based on snapshots. To give each transaction a consistent snapshot of the database at the moment of its commencement, MVCC produces several copies of a data item rather than obtaining locks. Because of this, read operations can continue without being blocked even when there are concurrent write operations.

OCC (Optimistic Concurrency Control): OCC gives optimistic assurances and operates under the premise that conflicts between transactions occur seldom. Without obtaining locks, transactions continue, and conflicts are examined at the commit phase. The impacted transaction is rolled back and restarted if conflicts are found. The blocking and costs associated with locking are reduced by OCC, but rigorous conflict detection and resolution procedures are needed.

Closing Remarks

In a multi-user database context, concurrency control is crucial for maintaining data integrity and making sure that transactions go through as intended. To coordinate transaction execution, avoid conflicts, and offer the required level of isolation and consistency guarantees, DBMS uses a variety of methods, including locking, MVCC, and OCC.

Top comments (0)