- Home
- Relational Databases
- CUBRID
CUBRID
What it is
CUBRID is a free, self-hosted object-relational database management system for Internet data services. Its architecture includes a database server, broker, and CUBRID Manager. Transactions support commit, rollback, and savepoints, with locking and logging used for consistency. Backup options cover online, offline, and incremental methods, while recovery can be partial or full. CUBRID HA synchronizes data across servers and can automatically fail over when a master node fails, although HA supports Linux only. Security features include TLS-encrypted client connections, documented access control and authorization, and transparent data encryption. Listed interfaces include CCI, JDBC, PHP, ODBC, OLE DB, ADO.NET, Ruby, Python, and Node.js. Downloads include CUBRID Manager and the CUBRID Migration Toolkit. The installation documentation lists Windows and 64-bit Linux, requiring at least 1 GB of memory and 2 GB of disk space. CUBRID supports read replicas and JSON data. The database engine is listed as free under Apache License 2.0, with APIs under the BSD license.
For and against
For
- Free self-hosted database engine
- Online, offline, and incremental backups
- Automatic failover through Linux HA
- Many language and database interfaces
- TLS and transparent data encryption
Against
- Managed service is not offered
- High availability supports Linux only
- Requires at least 1 GB memory
- Requires at least 2 GB disk space
The verdict
CUBRID offers a broad self-hosted database stack with transactions, backup, interfaces, and optional Linux high availability. Its operating-system and resource requirements should be checked before deployment.
Who it suits
CUBRID suits teams running their own database infrastructure on Windows or 64-bit Linux. It is relevant for applications needing transaction controls, varied language interfaces, backups, JSON data, or Linux high availability.
Plans and prices
CUBRID Database Engine
Database engine under Apache License 2.0; APIs under BSD license
Read 9 Oct 2026 · github.com
On the record
- Architecture
- CUBRID consists of a database server, a broker, and CUBRID Manager.cubrid.org · read 9 Oct 2026
- Backup and restore
- Supported backup methods include online, offline, and incremental backups; restore supports partial and full recovery.cubrid.org · read 9 Oct 2026
- Company location
- The CUBRID RDBMS GitHub organization lists its location as South Korea.github.com · read 9 Oct 2026
- Drivers and APIs
- The download documentation lists CCI, JDBC, PHP, ODBC, OLE DB, ADO.NET, Ruby, Python, and Node.js interfaces.cubrid.org · read 9 Oct 2026
- HA platform limit
- CUBRID HA supports Linux only.cubrid.org · read 9 Oct 2026
- High availability
- CUBRID HA synchronizes data across servers and can automatically fail over when a master node fails.cubrid.org · read 9 Oct 2026
- Management tools
- CUBRID downloads include CUBRID Manager and CUBRID Migration Toolkit.cubrid.org · read 9 Oct 2026
- Product
- CUBRID is an object-relational database management system optimized for Internet data services.cubrid.org · read 9 Oct 2026
- Security
- CUBRID supports TLS-encrypted client connections and documents access control, authorization, and transparent data encryption.cubrid.org · read 9 Oct 2026
- Supported operating systems
- The installation documentation lists Windows and 64-bit Linux as supported platforms.cubrid.org · read 9 Oct 2026
- System requirements
- The installation documentation specifies at least 1 GB of memory and 2 GB of disk space.cubrid.org · read 9 Oct 2026
- Transactions
- It supports transaction commit, rollback, and savepoints, and uses locking and logging to maintain transaction consistency.cubrid.org · read 9 Oct 2026
