MySQL

A relational database, focused on the internals of indexing, transactions and locking that underpin performance and correctness.

MySQL is one of the most widely used relational databases. Its InnoDB storage engine is known for its B+Tree indexes, MVCC and row-level locking. This subsection decomposes the internals behind performance tuning and correctness from two angles: index structure, and transactions and locks.

Articles

MySQL Indexing Internals and Best Practices — B+Tree, back-to-index lookups and covering indexes, the leftmost prefix rule and avoiding index invalidation.
Transactions and Locking — ACID, isolation levels, MVCC, row locks, gap locks and deadlock troubleshooting.

MySQL Indexing Internals and Best Practices

B+Tree structure, back-to-index lookups and covering indexes, the leftmost prefix rule, index selectivity and common anti-patterns that invalidate indexes.

Transactions and Locking

ACID and isolation levels, MVCC with undo/redo logs, row locks, gap locks and Next-Key Locks, plus deadlock causes and troubleshooting.