CUBRID 10.2 has been released.
Faster: CUBRID 10.2 has online parallel loader for faster bulk data loading which shows x10 performance to CUBRID 10.1, and improved auto-commit performance to improve YCSB (Workload A, B) workload by 130 to 140% compared to 10.1.
Better: CUBRID 10.2 supports built-in JSON data type and improved online parallel index creation and index management. In addition, the CTE (Common Table Expression) function has been extended so that it can be used with not only the syntax of SELECT but also INSERT, REPLACE, DELETE, UPDATE, CREATE TABLE AS SELECT etc.
Stronger & Stable: CUBRID 10.2 provides Double Write Buffer that protects against I/O partial writes, and has solved over 1300 issues, including many optimizations and refactoring to minimize performance degradation.
1. Key Features and Improvements
- Supports built-in JSON data type (including path, function, operator)
- Extended CTE function (INSERT, REPLACE, DELETE, UPDATE, CREATE TABLE AS SELECT also added)
- Improved auto-commit performance: 130 ~ 140% performance improvement in YCSB (A, B) test compared to 10.1
- Improved performance of bulk data import using online parallel load: 10 times better performance than 10.1
- Improved index performance and management
* supports online index creation without service impact
* supports parallel index creation
* improved foreign key generation performance
* supports invisible index syntax
- Release drivers and tools for 10.2
For more details, please refer to the CUBRID 10.2 online release notes (https://www.cubrid.org/manual/en/10.2/release_note/release_note_latest_ver.html) and online manuals (https://www.cubrid.org/manual/en/10.2/).
2. Precautions
- It is not compatible with the volumes of previous versions(10.1 or less), so DB reconfiguration is required when upgrading 10.2
3. Environmental requirements
- Linux: 64bit, kernel 2.4 or higher (Windows: Windows7 or higher)
- JRE 1.6 or higher to use JSP or CUBRID Manager
4. Download
- Product download: https://www.cubrid.org/downloads
- Product source code: https://github.com/CUBRID/cubrid/tree/v10.2
