Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Oracle Myths- Tablespace placement answered by Oracle TPC-C

Re: Oracle Myths- Tablespace placement answered by Oracle TPC-C

From: Pablo Sanchez <pablo_at_dev.null>
Date: Sun, 19 May 2002 06:56:22 -0600
Message-ID: <3ce79ef6_4@news.teranews.com>

"Andrew Mobbs" <andrewm_at_chiark.greenend.org.uk> wrote in message news:QpC*mZDop_at_news.chiark.greenend.org.uk...
> Pablo Sanchez <pablo_at_dev.null> wrote:
> >
> >I think the burden of proof is on folks who don't believe it's
> >important.
> >
> >As of this writing, the seventh place TPC-C result is an HP 9000
> >Superdome Enterprise Server running Oracle 9i. Reviewing the FDR,
on
> >page 98, we see Appendix E - Disk Storage. Notice that no table
> >segments are co-mingled on the same tablespace as any index/cluster
> >tablespaces:
>
> Thank you for providing the rope to hang yourself by :-)
>
> If you look at Table 5.2 of that document, you'll see that "Each
> logical volume is striped 74-way across all the arrays."

My bad. This is most excellent data that I missed. Thank you for pointing it out.

Given that RAID with cache performs drastically different than a regular disk ...

--
Pablo Sanchez, High-Performance Database Engineering
mailto:pablo_at_hpdbe.com
http://www.hpdbe.com
Available for short-term and long-term contracts
Received on Sun May 19 2002 - 07:56:22 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US