|
The actions behind the link are:
1) Downloaded JPOX plugin for commons-dbcp (jpox-dbcp-1.1.0-beta-3.jar) 2) Downloaded commons-dbcp (commons-dbcp-1.2.1.jar) 3) Downloaded commons-pool (commons-pool-1.2.jar) 4) Downloaded commons-collections (commons-collections-3.1.jar) 5) Added a JPOX property to PMF properties (org.jpox.connectionPoolingType=DBCP) 6) Added all jars to the classpath and started the TCK However, connection pooling does not speed up TCK execution: Each test case creates a new PMF instance and the connection pool is bound to a PMF instance. The performance would improve, if many test cases performed multiple transaction. Actually, most of them performs a single transaction. I propose to use "c3p0-0.9.0-pre6.jar" for c3p0. This is the current pre release 6. A TCK run with this version produces the same amount opf errors as without connection pooling:
Datastore identity: Tests run: 350, Failures: 13, Errors: 88 (Time: 625 seconds) Application identity: Tests run: 350, Failures: 12, Errors: 84 (Time: 768 seconds) The download URL is: http://prdownloads.sourceforge.net/c3p0/c3p0-0.9.0-pre6.bin.zip?use_mirror=switch Reopened to set the Fix Version/s field to JDO 2 beta.
|
||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
Improvement
Closed
Major
Use JPOX connection pooling to improve performance
"One thing that you may find beneficial when running the TCK is to enable use of connection pooling via the JPOX plugin for Apache DBCP, or C3P0. Speed-ups of 30-40% are normal. See here for how to enable it
http://www.jpox.org/docs/1_1/rdbms_connection_pooling.html"