benchmarkingblog

Benchmarking and Systems Performance

Oracle’s SPECjEnterprise: Tricks and Tomatoes

with one comment

There are so many important and unpleasant things going on in the world today. I know for sure that I definitely want to avoid the following words: stock market, war, economy, London. I’d rather discuss a subject I might actually be able to do something about. So today I am going to talk about squirrels.

You see, the war is over and the squirrels have won.

They’ve been through hot pepper spray, chicken wire, an automated sensor sprinkler, transportation of four to a park, and coffee grounds. And they’ve still made off with almost every green tomato we have.

Last year the dog ate some of them. But the squirrels are so much more cunning. They seem to know every trick in the book.

And that’s exactly how I feel about Oracle’s newest press release published yesterday on the SPECjEnterprise2010 Java benchmark.

  • What they don’t tell you is that their new “record-setting” blade result on this benchmark actually used twice the memory on the app server as the IBM blade result.
  • What they don’t tell you is that their new blade result used much more storage than the IBM blade result.
  • What they don’t tell you is that they rounded up, not down, when computing their performance claim. Should a true leader need to use the word “nearly”?
  • What they don’t tell you is that in other claims in the press release they compare multiple node to single node results.
  • What they don’t tell you is that they reference not just one but two SPEC Java RETIRED benchmarks.
  • What they don’t tell you is that one of “their” results is not even on their system – unless they just bought Dell.

Four squirrels in the park later, I knew we had them beat. We actually had two days where I didn’t see any in the garden. Or any in our whole yard, for that matter. And then, wouldn’t you know it, another one comes scurrying along the fence, a nice green tomato in hand. Going back to his home in the big tree — in my neighbor’s yard.

************************************************

Source: http://www.spec.org

SPEC, SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPECjEnterprise, SPECjvm, SPECvirt, SPECompM, SPECompL, SPECsfs, SPECpower, SPEC MPI and SPECpower_ssj are trademarks of the Standard Performance Evaluation Corporation (SPEC).

The postings on this site solely reflect the personal views of the author and do not necessarily represent the views, positions, strategies or opinions of IBM or IBM management.

technorati tags: , , , , ,,,,,,,,,,,,,

About these ads

Written by benchmarkingblog

August 12, 2011 at 12:40 pm

One Response

Subscribe to comments with RSS.

  1. What do you mean by ” they reference not just one but two SPEC Java RETIRED “

    noname

    February 8, 2012 at 9:00 am


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: