Overview
- 1MySQL is an open source relational database management system originally built by MySQL AB, acquired by Sun Microsystems in 2008 and then by Oracle in 2010.
- 2It ships with the InnoDB storage engine by default, providing ACID-compliant transactions, row-level locking, and crash-safe recovery.
- 3Oracle now ships MySQL on a dual-track model: 8.4 LTS for long-term production stability and 9.x Innovation releases every few months for early access to new features.
- 4Released under the GPL v2 license with a commercial dual-licensing option, MySQL is free for most use cases and forms the "M" in the classic LAMP/LEMP stack.
- 5It powers a huge share of the web, from WordPress and Drupal sites to large-scale platforms that shard MySQL horizontally with middleware like Vitess.
Key Features in 8.4 LTS
InnoDB storage engine with ACID transactions, MVCC, foreign keys, and clustered primary-key indexes.
Native JSON data type with dedicated functions and generated columns for indexing values inside JSON documents.
Window functions and recursive common table expressions (CTEs) for analytical queries, available since MySQL 8.0.
Multiple replication topologies: asynchronous, semisynchronous, and GTID-based, plus Group Replication/InnoDB Cluster for automated failover.
Pluggable storage engine architecture (InnoDB, MyISAM, Memory, Archive) letting different tables use different storage strategies.
Built-in security features including roles, caching_sha2_password authentication, transparent data encryption, and fine-grained privilege management.
Instant DDL operations (such as adding columns) and online schema changes that avoid locking large tables during routine maintenance.
Use Cases
- →Backend database for content management systems and web apps (WordPress, Drupal, Magento) in classic LAMP/LEMP deployments.
- →OLTP workloads for e-commerce, SaaS, and fintech applications that need reliable ACID transactions at moderate to large scale.
- →Horizontally sharded, globally distributed deployments using Vitess (the sharding layer behind PlanetScale and YouTube) or ProxySQL for pooling and routing.
- →Managed cloud deployments via Amazon RDS/Aurora, Google Cloud SQL, and Azure Database for MySQL for teams that want automated backups, patching, and scaling.
What's New in MySQL 8.4 LTS & 9.x
- MySQL 8.4 (April 2024) became the first release under Oracle's new Innovation/LTS model, offering years of support as the stable production baseline succeeding 8.0.
- MySQL 9.x Innovation releases ship every few months, introducing a native VECTOR data type for storing and querying embeddings in AI/similarity-search workloads.
- A new HyperGraph-based query optimizer is available (opt-in) in recent releases, improving execution plans for complex multi-way joins.
- Innovation releases carry a much shorter support window than LTS and are aimed at teams that want the newest features before they land in the next LTS.
- MySQL 8.0 is winding down toward end of standard Oracle support, pushing most production users toward 8.4 LTS as the recommended upgrade target.
- Continued refinements to redo log and undo log handling improve write throughput and crash recovery time on busy InnoDB instances.
- Major cloud vendors have rolled out managed 8.4 LTS offerings, cementing it as the default choice for new managed-database deployments.
Core SQL Features
- Window functions (ROW_NUMBER, RANK, LAG/LEAD, moving aggregates) available since MySQL 8.0 for analytical queries without self-joins.
- Recursive and non-recursive common table expressions (WITH clauses) for hierarchical and multi-step queries.
- Native JSON column type with JSON_EXTRACT, JSON_TABLE, and generated/virtual columns for indexing values inside JSON documents.
- Descending indexes and invisible indexes, letting DBAs test the impact of dropping an index before actually removing it.
- CHECK constraints (enforced since 8.0.16) combined with CTEs for declarative, set-based data validation.
- Full-text search indexes and spatial data types/functions for geometry-based queries alongside standard relational data.
Performance Tuning
- InnoDB buffer pool sizing is the single biggest lever for read performance; it should typically hold the working set in memory.
- The Performance Schema and sys schema expose wait events, query latency, and lock contention for targeted tuning.
- EXPLAIN ANALYZE (added in 8.0.18) shows actual execution timings per plan node instead of just estimated costs.
- Proper indexing strategy — covering indexes and composite index column order — remains the most common fix for slow queries.
- Instant ADD COLUMN and other online DDL operations reduce downtime for schema migrations on large InnoDB tables.
- Connection pooling via ProxySQL or the application layer avoids overhead from MySQL's thread-per-connection model under high concurrency.
Replication & Ecosystem
- Asynchronous, semisynchronous, and GTID-based replication support a range of consistency and durability trade-offs for read replicas and disaster recovery.
- MySQL InnoDB Cluster/Group Replication provides built-in multi-primary or single-primary HA with automatic failover, orchestrated via MySQL Router.
- MariaDB, forked from MySQL in 2009 after Oracle's acquisition of Sun, remains a widely used alternative with its own storage engines and release cadence.
- Percona Server and Percona XtraDB Cluster offer open source, enhanced builds with additional performance and monitoring tooling.
- Vitess, the sharding middleware originally built at YouTube, powers PlanetScale and lets MySQL scale horizontally across many nodes.
- Managed offerings — Amazon RDS/Aurora MySQL-compatible, Google Cloud SQL, Azure Database for MySQL — handle backups, patching, and read replicas for teams that don't want to self-host.
- A large ecosystem of GUI and CLI tools (MySQL Workbench, phpMyAdmin, DBeaver) and drivers for every major language keeps MySQL accessible to developers of all levels.
Frequently Asked Questions
MySQL vs PostgreSQL — which should I choose?▼
MySQL is generally simpler to set up and tune for straightforward read-heavy web and OLTP workloads, and it has an enormous hosting and tooling ecosystem built around WordPress and LAMP stacks. PostgreSQL tends to be preferred for complex queries, strict standards compliance, and advanced data types like JSONB or extensions such as PostGIS and pgvector. Both are mature, ACID-compliant, and battle-tested at scale, so the choice often comes down to ecosystem fit rather than raw performance.
What is InnoDB and why is it MySQL's default storage engine?▼
InnoDB is MySQL's default transactional storage engine, providing ACID compliance, row-level locking, multi-version concurrency control (MVCC), and crash-safe recovery via redo and undo logs. It stores each table as a clustered index organized around the primary key, which makes primary-key lookups very fast. InnoDB replaced MyISAM as the default engine back in MySQL 5.5 specifically because MyISAM lacks transactions and foreign keys.
Is MySQL free to use commercially?▼
Yes. MySQL Community Edition is released under the GPL v2 license and can be used, embedded, and deployed commercially at no cost, which is why it underpins most WordPress sites and LAMP-stack applications. Oracle also sells MySQL Enterprise Edition, which adds paid tools like enterprise backup, firewall, and audit features, plus a commercial license for vendors who need to embed MySQL in proprietary software without GPL obligations.
What's the difference between MySQL 8.4 LTS and the 9.x Innovation releases?▼
MySQL 8.4 is the Long-Term Support release, receiving years of bug and security fixes and recommended as the stable baseline for production. The 9.x line is the Innovation track, shipping new features such as the VECTOR data type and a HyperGraph query optimizer every few months, but each release gets only a short support window before the next one supersedes it. Most production teams should run 8.4 LTS unless they specifically need a 9.x feature.
How does MySQL replication and high availability work?▼
MySQL supports asynchronous and semisynchronous replication out of the box, using GTIDs (global transaction identifiers) to track and reapply transactions on replicas reliably. For automated failover, MySQL InnoDB Cluster combines Group Replication with MySQL Router to provide single- or multi-primary HA without a separate third-party tool. For horizontal scaling beyond one primary's write capacity, teams add sharding middleware like Vitess.
How is MariaDB related to MySQL, and should I use it instead?▼
MariaDB is a community fork of MySQL created in 2009 by MySQL's original developer, Monty Widenius, in response to Oracle's acquisition of Sun Microsystems, MySQL's owner at the time. It started as a close drop-in replacement but has since diverged with its own storage engines (such as Aria and ColumnStore) and features, so compatibility with recent MySQL versions isn't guaranteed. Choosing between them usually comes down to hosting support and specific features rather than one being strictly better than the other.