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: New IBM Nonsense

Re: New IBM Nonsense

From: Serge Rielau <srielau_at_ca.ibm.com>
Date: Sun, 23 Jan 2005 09:23:53 -0500
Message-ID: <35hqamF4djb4oU1@individual.net>


thu.nnguyen_at_gmail.com wrote:
> Mark Townsend wrote:
>

>>thu.nnguyen_at_gmail.com wrote:
>>
>>
>>>
>>>There is a TPC-C result running on identical Power5 hardware (8 way

>
> p5
>
>>>570).  You can check it out at www.tpc.org. It shows DB2 result is
>>>~430k and Oracle is at ~370k tansactions /minute.  DB2

>
> significantly
>
>>>faster.
>>>
>>
>>
>>You may want to look at the differences in the disk environment

>
> between
>
>>the two 'apples to apples' configurations. The IBM result used at

>
> least
>
>>  20% more spindles than the Oracle result. It also had at least 30%
>>more clients driving the backend. This may explain why the IBM result

>
>
>>was 15% faster.

>
>
> What are you driving at? IBM is delibately crippling the Oracle result?
> Somehow I doubt it. I'm guessing its just a case of DB2 being more
> optimized and therefore faster for the AIX/Power5 system. As for more
> clients/spindle .. well if the system can handle it, they would need
> more clients/spindles. Adding more spindles/clients to a system
> already running at near 100% CPU load would not get you a higher TPC-C
> result.
>

Can't comment on Oracle's strategy, but for the high end results (in contrast to those geared towards low price/performance). DB2 engineers folks provide whatever the system can digest. The goal of the tests is not to minimize spindles.
If Oracle publishes numbers that held back IO, too bad.

Cheers
Serge

-- 
Serge Rielau
DB2 SQL Compiler Development
IBM Toronto Lab
Received on Sun Jan 23 2005 - 08:23:53 CST

Original text of this message

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