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

Home -> Community -> Mailing Lists -> Oracle-L -> Re: SPARC vs. RS/6000 - which has better performance/processor?

Re: SPARC vs. RS/6000 - which has better performance/processor?

From: Jared Still <jkstill_at_cybcon.com>
Date: Thu, 10 May 2001 17:30:46 -0700
Message-ID: <F001.0030057E.20010510173030@fatcity.com>

On Thursday 10 May 2001 11:16, Chris Rezek wrote:
> We are looking at new servers for our Oracle8i OLTP back-end. An IBM
> sales guy claimed that an RS/6000 is 'twice as fast' per processor
> running Oracle8i in an OLTP environment - a 6 processor RS/6000 is just
> as fast as a 12 processor SPARC.
>
> What do y'all think about this?

Funny you should ask.

I wrote a paper at a previous employer comparing RS/6000 and Sun boxes based on throughput. This did not take into consideration the storage medium, whether SSA's, SAN or whatever.

This was strictly to compare the abilitity of the machines to process data.

I started by building some Perl tools that collected data at the OS level from iostat, vmstat and mpstat, and saved it to tables with DBD::Oracle.

On the database side there was some PL/SQL I put together to collect Oracle performance data.

These tools were run for a few weeks on our production database box; Orale 7.3.4 on Solaris 2.6.

I then collected all the performance data that could be found regarding benchmarks on RS/6000's and Sun's. I know how many of you feel about benchmarks, and I agree, but this is all I had to work with.

Using the benchmark data and collected data I was able to make some assumptions about the scalability and performance of the two architectures, and the data from all this was then plugged into an Excel workbook that was setup to use queuing theory to predict performance.

This Excel workbook is available at Craig Shallahamer's excellent website, www.orapub.com. It is used as a tool in his capacity planning class. It was a very educational class.

The result of this? If you're looking at data processing ability, then I would agree with the IBM rep. Actually, I found that the RS/6000 CPU's were about 2.3 times faster than Sun's.

A 16 way IBM S80 is easily the equivalent of a Sun 36 way E10000, and quite a bit less money to boot.

That said, 2 things come to mind.

  1. The DH Brown report rates Solaris 8 much higher than AIX as an enterprise operating systenm, a situation that is reversed from previous years.

http://www.dhbrown.com/dhbrown/OpSys.cfm

2. I personally much prefer Solaris to AIX.

Jared

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Jared Still
  INET: jkstill_at_cybcon.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).
Received on Thu May 10 2001 - 19:30:46 CDT

Original text of this message

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