Re: "db file async I/O submit" wait event : 11gR2

From: Dirk Drexler <dirk.drexler_at_gmx.de>
Date: Fri, 30 Jul 2010 07:21:48 +0200
Message-ID: <20100730052148.240310_at_gmx.net>



Hi Aditya,

without seing your data...this could be an issue with fragmented data. Try to reorg the tables/indexes with:
alter table <tablename> move;
Please notice that the rowids will be changed so that you have to rebuild the indexes afterwards with: alter index <indexname> rebuild

Greetings Dirk

  • Original-Nachricht --------
    > Datum: Thu, 29 Jul 2010 13:13:17 -0700
    > Von: Yadavalli Aditya <aditya.yadavalli_at_gmail.com>
    > An: tim_at_evdbt.com, martin.a.berger_at_gmail.com, dirk.drexler_at_gmx.de
    > CC: Oracle-L_at_freelists.org
    > Betreff: Re: "db file async I/O submit" wait event : 11gR2

> Thank you for your help and time.
>
> I have observed more time being spent on checkpointing after
> increasing redo logs to 500M, so reduced them to 250m and
> filesystemio_options (changed to setall from async).
>
> After that insert performance is really improved, need to check
> further on Update and Delete performance (They are still not as
> expected). Have to work on more priority tasks so setting it aside for
> some time. Will update the group soon on further progress.
>
> Thanks again.
> -Aditya.

--
http://www.freelists.org/webpage/oracle-l
Received on Fri Jul 30 2010 - 00:21:48 CDT

Original text of this message