What are the types of anomalies and which levels of isolation address them?

In the context of databases and transactions, the following main anomalies are distinguished during concurrent execution:

• Dirty Read: a transaction reads data that has been modified but not yet committed by another transaction.
• Non-repeatable Read: when reading the same data multiple times within a transaction, it may change due to another transaction.
• Phantom Read: when re-executing a query within a transaction, new rows added by another transaction appear.

Isolation levels that address these anomalies:

• Read Uncommitted — does not prevent any anomalies (dirty reads are possible).
• Read Committed — prevents dirty reads but allows non-repeatable and phantom reads.
• Repeatable Read — prevents dirty and non-repeatable reads, but phantom reads are possible.
• Serializable — the strictest level, prevents all listed anomalies, ensuring full transaction isolation.

Thus, the choice of isolation level depends on the system's requirements for consistency and performance.