[jira] Updated: (OPENJPA-342) Performance of flush()

["Christiaan (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/OPENJPA-342?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Christiaan updated OPENJPA-342:
-------------------------------

    Attachment: TestCaseFlushPerformance.ZIP

Testcase for reproducing the performance results.
Two variables are used:
generateLargeDataset : should be set once for generating a large dataset
singleTransaction : to switch between the two modes (one large transaction, or multiple small transactions)relevant for this performance testcase

Included "Performance results flush.txt" to see the difference of the two modes

> Performance of flush()
> ----------------------
>
>                 Key: OPENJPA-342
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-342
>             Project: OpenJPA
>          Issue Type: Bug
>         Environment: Windows XP, jdk1.6.0_02, Kodo 4.1.4, Ms sql server 2005, jTDS 1.2 database driver
>            Reporter: Christiaan
>         Attachments: TestCaseFlushPerformance.ZIP
>
>
> Based on the discussion in thread http://www.nabble.com/Performance%3A-single-vs-multiple-transactions-tf4272115.html
> I've created a testcase, which is attached. It includes a "Performance results flush.txt" which gives an overview of the the performance results. In general, using flush() within one transaction with many changes the performance drops quite significantly (x 15). 
> Btw, not sure which OpenJPA version this is. It is based on Kodo jdo version 4.1.4
> regards,
> Christiaan

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.