categories.database Advanced

What Are the Four Transaction Isolation Levels and What Problems Do They Solve?

AI Practice

Transaction Isolation Levels

Four Anomaly Types

  • Dirty Read: Reading uncommitted data from another transaction
  • Non-repeatable Read: Reading the same row twice in one transaction yields different results (another transaction updated it in between)
  • Phantom Read: Two queries in one transaction return different row counts (another transaction inserted/deleted rows)
  • Lost Update: Two concurrent updates to the same row; one overwrites the other

Four Isolation Levels (low to high)

Level Dirty Read Non-repeatable Read Phantom Read
Read Uncommitted Possible Possible Possible
Read Committed Prevented Possible Possible
Repeatable Read Prevented Prevented Possible
Serializable Prevented Prevented Prevented

Performance Trade-off

Higher isolation = lower concurrency performance (more locking)

Default Levels

  • MySQL (InnoDB): Repeatable Read
  • PostgreSQL: Read Committed
  • Oracle: Read Committed

Interview bonus: PostgreSQL uses MVCC (Multi-Version Concurrency Control) for isolation without locks—readers don't block writers, giving better performance than traditional locking.

✦ AI Mock Interview

Type your answer and get instant AI feedback

Sign in to use AI scoring

Copyright © 2026 Wood All Rights Reserved · FE Interview Hub