Re: IBM DB2 Press Release, hogwash!?

nsouto_at_nsw.bigpond.net.au.nospam (Nuno Souto) wrote:
>On Sun, 30 Jul 2000 05:06:22 GMT, cosmin_ioan_at_my-deja.com wrote:
>
>>I happen to have stumbled upon a news article of IBM, at
>>http://biz.yahoo.com/bw/000729/ny_ibm_sof.html and am very
 concerned
>>of the statements there, particularly,
>
>I wouldn't worry too much about these company press releases.
>Note that the text search technology is touted as being break-
 through
>and fatser than a speeding bullet, then no details are given on
>exactly how does this help the benchmark performance. Typical.
>
>>
>>Is this true? and if so, how does Oracle plan to respond?
>>
>
>Don't worry, they';ll respond. Last time I looked, nobody
 could make
>code run faster than the CPU can run, so ORACLE can also match
 these
>wild figures.
>
>
>Cheers
>Nuno Souto
>nsouto_at_nsw.bigpond.net.au.nospam
>http://www.users.bigpond.net.au/the_Den/index.html
>

What the press release is refering to is the new TPC-C bench mark that IBM' DB2 set running on an IBM Netinfinity box under the MS NT OS. These figures surpase the MS-Server numbers that Microsoft claimed but then had to withdraw because the test was found not to conform to the standard. I believe that the code necessary to support key updates was missing, though MS refused to comment on what effect this would have had on the performance results.

The numbers refering to cost have to do with the hardware/software cost to the transaction rate. The TPC or is it TCP -C test is for distributed transactions and given a choice of 8 2-cpu boxes or 1 16-cpu box most shops would choose the a 16 cpu smp machine over a cluster. Still I would like to see what Oracle can do running on the same setup. Those Netinfinity boxes use IBM copper technology which runs faster than the existing Aluminium technology. IBM says it is cheaper and easier to manufacuture from a defect point of view than the existing technology.

I believe that Oracle and Sun have both probably taken note of these numbers, but I am sure they will try to wave them off as not being that significant. Perhaps this will help make Oracle straighten out its pricing as it is nearly immpossible to get consistent quotes from Oracle and they are often unrealisticly high.

Most of the information in this post is from a Computer World article on the MS bench mark withdrawal. Obviously a couple of the comments are just my opinion.

• Mark D. Powell -- The only advice that counts is the advice that you follow so follow your own advice. --

  ---

Got questions? Get answers over the phone at Keen.com. Up to 100 minutes free!
http://www.keen.com Received on Sun Jul 30 2000 - 00:00:00 CDT