Multiversion Based Techniques
Multiversion Based Techniques
Multiversion Based Techniques
• Read-Consistency Models
• Snapshot Isolation
• Conclusion
These techniques maintain multiple versions of data items to enable concurrent read and write operations.
Multiversion Concurrency Control: Technique that maintains multiple versions of data items.
Each transaction reads a consistent snapshot of the database, ensuring concurrency and isolation.
Readers do not block writers, and writers do not block readers, allowing high concurrency.
• Read Committed: Each transaction reads the most recent committed version of a data item.
• Repeatable Read: Each transaction reads a consistent snapshot of the database, preserving the consistency
of read data items throughout the transaction.
• Serializable: Each transaction sees a consistent snapshot of the database and guarantees serializability.
Snapshot Isolation:
Each transaction reads a consistent snapshot of the database at its start time.
Transactions operate on their snapshot and do not block or interfere with each other.
Write conflicts are resolved by aborting the younger transaction, ensuring data consistency.
Improved Concurrency: Multiple transactions can read and write data items simultaneously without blocking
each other.
Read-Consistency: Provides consistent snapshots for read operations, preserving data integrity.
Increased Storage Overhead: Maintaining multiple versions of data items requires additional storage space.
Visibility of Uncommitted Data: Read operations may access uncommitted data, leading to anomalies if not
properly managed.
Constraint Validation: Ensuring consistency and enforcing integrity constraints across multiple versions can
be challenging.
Data Base Management Systems
Conclusion
MVCC, read-consistency models, and multiversion timestamp ordering enable concurrent read and write
operations.