What a storage engine is
MySQL separates the SQL layer that parses and optimizes queries from the storage engine that actually stores rows and serves reads and writes. InnoDB has been the default engine since MySQL 5.5 and is the right choice for almost every workload.
What InnoDB gives you
- ACID transactions with commit and rollback, backed by redo and undo logs.
- Row level locking so two sessions writing different rows do not block each other.
- Crash recovery that replays committed work after an unexpected shutdown.
- Foreign keys for referential integrity, which the older MyISAM engine never supported.
How it stores data
InnoDB keeps table data and indexes in B plus tree structures on disk and caches hot pages in a memory area called the buffer pool. Rows live inside a primary key ordered tree, so the physical layout follows the primary key.
Older engines like MyISAM offered fast counts and table level locks but no transactions or crash safety, which is why InnoDB replaced them for general use.
Key idea
InnoDB is MySQL's default transactional engine, giving ACID transactions, row level locking, and crash recovery on top of B plus tree storage and an in memory buffer pool.