These advanced DBMS concepts show up often in backend system design discussions.

MVCC

Multi-Version Concurrency Control lets readers and writers work concurrently by keeping multiple row versions.

Readers can see a consistent snapshot while writers update newer versions.

WAL

Write-Ahead Log records changes before applying them to data files.

If the DB crashes, WAL helps recover committed changes.

Replication

Copies data from one database node to another.

Common forms:

• Primary-replica.
• Multi-primary.
• Synchronous.
• Asynchronous.

CAP Theorem

In a network partition, a distributed system must choose between:

• Consistency.
• Availability.

Partition tolerance is unavoidable in distributed systems.

BASE

Often discussed with distributed NoSQL systems:

• Basically Available.
• Soft state.
• Eventually consistent.

Connection Pooling

Opening DB connections is expensive. A pool reuses connections.

Benefits:

• Lower latency.
• Controlled DB connection count.
• Better resource usage.

Redis Caching

Redis is often used for:

• Cache.
• Sessions.
• Rate limits.
• Queues.
• Counters.

Hard parts:

• Cache invalidation.
• Stale data.
• Choosing TTLs.
• Avoiding cache stampedes.

Interview Notes

• MVCC improves read/write concurrency.
• WAL supports durability and crash recovery.
• Replication improves read scale and availability.
• CAP explains tradeoffs during network partitions.
• Connection pools protect both app and database resources.