Last checked and updated on 05 Jun 2026
What Is New in MySQL 5.6
MySQL 5.6 delivers major enhancements in performance, scalability, and manageability. This release focuses on making the database faster and more reliable for modern web, cloud, and embedded use cases.
| Category | Key Changes |
|---|---|
| Performance & Scalability | InnoDB improvements, Optimizer enhancements, Better replication |
| Manageability | Online DDL, PERFORMANCE_SCHEMA expansions, New instrumentation |
| Security | Password policies, SHA-256 authentication, Security-sensitive logs |
| New Features | Memcached API, Full-text search in InnoDB, TIMESTAMP auto-update |
| Deprecated/Removed | Deprecated parameters, Removal of old partitioning options |
How did performance and scalability improve?
The InnoDB storage engine got a significant rewrite, making it the clear choice for most workloads. Subquery optimizations and a new optimizer cost model drastically cut query execution times.
Binary log group commits reduce replication overhead, and multi-threaded slaves speed up applying changes from the master. You can now run ALTER TABLE without locking out your application for many operations.
Key Optimizer Changes
- Subquery optimizations using semi-join and materialization
- Index Condition Pushdown (ICP) for faster range scans
- Batch Key Access (BKA) to reduce random disk I/O for joins
What new management tools were introduced?
Online Data Definition Language (DDL) is a game-changer. You can add indexes, change columns, and optimize tables without taking the database offline or blocking writes.
The PERFORMANCE_SCHEMA was expanded with dozens of new tables. It now gives you deep visibility into statement execution, table I/O, and user activity. New EXPLAIN formats make it easier to understand query plans.
New Instrumentation
- Tables for monitoring table and index usage
- Statement digest tables for identifying frequent queries
- User and host-based activity statistics
Were there any security enhancements?
Yes, MySQL 5.6 tightened up security with more granular controls. Password validation policies help enforce stronger passwords, and you can now use SHA-256 based authentication for more secure password hashing.
Security-sensitive information is now automatically redacted from the plaintext logs. This prevents accidentally logging passwords or other sensitive data from SQL statements.
What entirely new features were added?
The integrated Memcached API lets you use InnoDB tables as a key-value store, bypassing the SQL layer for ultra-fast reads and writes. This is huge for caching and simple data retrieval patterns.
Full-text search (FTS) support finally came to InnoDB, so you no longer need to choose between transactions and text search. The TIMESTAMP and DATETIME types also got auto-initialization and update properties.
CREATE TABLE example (
id INT PRIMARY KEY,
data TEXT,
modified TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP
);FAQ
Is the Memcached API a replacement for a dedicated Memcached server?
Not exactly. It provides a NoSQL interface to your InnoDB data, offering persistence and consistency. It's great for reducing SQL overhead on simple lookups, but for pure, transient caching, a dedicated cache server might still be faster.
How does online DDL actually work? Are there any catches?
It works by using internal optimizations in InnoDB to avoid table copying and metadata locking for specific operations. The main catch is that some complex operations, like changing the data type of a column, still require a table copy and lock.
Should I always use the new cost-based optimizer?
In most cases, yes. The new cost model is smarter and uses statistics about your data distribution. However, if you have a very stable query plan from 5.5 that you rely on, test thoroughly as the new optimizer might choose a different path.
What's the biggest reason to enable PERFORMANCE_SCHEMA?
To finally figure out what your database is actually spending its time doing. It moves you from guessing about performance bottlenecks to having concrete data on wait events, query latency, and resource consumption.
Does full-text search in InnoDB have all the same features as MyISAM FTS?
Mostly, but not entirely. The core functionality is there, but some more advanced features like proximity operators and finer control over the stopword list were still maturing in this initial release.