Faster: CUBRID 10.1 has almost doubled the optimum workloads of TPC-C benchmark compared to CUBRID 9.3.
We also followed and improved results for the maximum tpmC by about 80%. in addition, Sysbench improved 30%, YCSB (Yahoo! Cloud Serving Benchmark) improved performance by 50-65%, and in the case of TPC-W, Shopping Mix improved 6 times and Ordering Mix improved 4 times.
Better: CUBRID 10.1 has new features such as CTE (Common Table Expression). Developers can write complex queries easier and clearer, including recursive queries.
Stronger & Stable: CUBRID 10.1 is made even more powerful by rewriting the entire storage module. Also, around 5,000 product issues have been fixed in CUBRID 10.0 and CUBRID 10.1 alone, which is more than the number of issues resolved in all 9.x versions. In addition, the stability of HA has been improved even under heavy workloads, and replication delay has been significantly reduced.
1. Key Features and Improvements
- Supports snapshot isolation level(performance improvement of 10.0)
- Adds COMMENT function to DDL syntax(table/column/view/index/user/serial/trigger/sp etc.)
- Supports CTE (used in the WITH clause) query
- Supports default to_char (sysdate|systime|sysdatetime|systimestamp, format)
- Supports arithmetic operations in the LIMIT clause
- Supports timezone and functions related to date time and timestamp data types
- Supports compression and storage of string type data
For more details, please refer to the CUBRID 10.1 online release notes(English) and online manuals(English), and Korean materials will be provided soon.
- It is not compatible with the volumes of previous versions(10.0, 9.3 or less), so DB reconfiguration is required when upgrading 10.1
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
- Product download: https://www.cubrid.org/downloads
- Source code download: https://github.com/CUBRID/cubrid/tree/v10.1