[jboss-jira] [JBoss JIRA] Commented: (EJBTHREE-649) Query Performance within transaction decreasing progressively when using default FlushModeType.AUTO

Thu Sep 21 20:31:01 EDT 2006

    [ http://jira.jboss.com/jira/browse/EJBTHREE-649?page=comments#action_12343794 ] 

Emmanuel Bernard commented on EJBTHREE-649:
-------------------------------------------

We have discovered an important unefficiency (n**2 class) in HEM flush algorithm, do you happen to have lot's of cascade (even bidirectional ones)?

> Query Performance within transaction decreasing progressively when using default FlushModeType.AUTO
> ---------------------------------------------------------------------------------------------------
>
>                 Key: EJBTHREE-649
>                 URL: http://jira.jboss.com/jira/browse/EJBTHREE-649
>             Project: EJB 3.0
>          Issue Type: Bug
>    Affects Versions: EJB 3.0 RC8 - FD
>         Environment: WinXP SP2, Oracle 9i Enterprise Rel. 9.2.0.1.0, JBoss 4.0.4GA, EJB3.0 RC8-FD
>            Reporter: Andreas Zimmer
>         Assigned To: Emmanuel Bernard
>
> Query Performance within a Transaction is progressively decreasing when working with EJB3.0 default FlushModeType.AUTO:
> 20 queries/reads within transaction - 10 msecs average per query/read 
> 1000 (same as before) queries/reads within transaction - 130 msecs average per query/read (factor 13+) 
> With FlushModeType.COMMIT query performance within the same scenario ist constant at expected 10 msecs but FlushModeType.COMMIT imposes restrictions which our projects can't cope with. We need updates within transaction being visible in subsequent queries of the same transaction.
> The case is explained in some detail in the JBoss Forum Thread as provided in JBoss Forum Reference.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira