Ramon F Herrera wrote:
> Daniel Morgan <damorgan_at_x.washington.edu> wrote in message
>
>>Not my experience at all. I'm seeing 20-30% increases in almost
>>everything I do.
>
>
> Well, maybe I am comparing apples and oranges. See my points
> of comparison below. The older machine is still in production
> and the newer one is under "observation" because I am less
> than thrilled with its current performance. I both cases I
> used Werner Puschitz' instructions to the letter, as far
> as kernel configuration, installation, etc.
>
> Old Oracle Server:
> Dell PowerEdge 2550 with 1 2.4 GHz CPU and 2.5 GB RAM
> PERC RAID chip
> RedHat 9
> Oracle 9iR2
>
>
> New Oracle Server:
> Dell PowerEdge 2650 with 2 3.0 GHz CPUs and 2 GB RAM
> Adaptec RAID card with full cache
> RedHat Enterprise Linux 3
> Oracle 10gR1
>
>
> The new server is substantially better in every respect
> -except for having less RAM. Could this be what's slowing it down?
> I see no swapping at all.
>
> A typical daily job that takes 7-8 minutes in the old machine,
> takes 40 minutes in the new one. The old -and presumably less
> capable- server is up to ten times faster for some jobs.
> I have tested my own SQL procedures and basic Oracle tasks
> such as indx creation. The old guy always beats the new one.
>
> I understand that RedHat and Oracle did a lot of work in
> order to improve Linux and produce the RHEL, so it should
> be much better matched to the Oracle database needs the RH9.
>
> Any ideas, suggestions?
>
> -Ramon
What NetComrade says. But make sure you have current statistics
generated with DBMS_STATS. Make sure you run DBMS_XPLAN to get
explain plans for comparison, etc.
Running 9i and 10g on the exact same hardware I have always seen
a measurable performance improvement with 10g over 9i.
--
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)
Received on Tue Mar 16 2004 - 15:22:20 CST