Benchmarking and systems performance - hosted by Elisabeth Stahl
Autores do Blog: ELISABETH STAHL 0600029JQ8 is Chief Technical Strategist and Executive IT Specialist, IBM Systems and Technology Group, and has been working in systems performance for over 25 years.
See what happens when Oracle compares SAP SD with SAP SD-Parallel benchmark results . . .
I read last week about an exciting challenge to limit yourself to wearing only six items of clothing in one month. A very cool exercise to demonstrate how little we really need to consume. Intriguing yes. I was immediately hooked into thinking about which items I would choose. Until I read that in this "shopping diet" you could actually use multiples of the same item if they were similar. So if I had three pairs of black jeans, I could count that as one. Easy to stack the deck here, especially if you like black.
A few weeks ago Oracle published an SAP SD-Parallel benchmark result. And a press release about it. One could question whether it's really news when someone touts a .4% difference. Yes, that's point four percent. But that's for another story.
The point here is that Oracle compared this parallel SD result to other vendors' regular SD results. Which is not illegal. But it could be stacking the deck. Yes, it's true, the two benchmarks use the same business transactions and processes. But there is a big difference. Parallel benchmarks employ data distribution. As highlighted on the SAP website “It is generally accepted that data distribution can significantly influence the benchmark result in a parallel environment.”
The benchmarks clearly have different names -- SD vs. SD-Parallel -- for a reason. So the difference will be recognized. But possibly only by the reader who is looking for these things.
Which reminded me of the husband of the woman who only wore six items. He didn't even notice. She was upset because he's the one who does the laundry -- and he didn't even notice.
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.
SAP, mySAP and other SAP product and service names mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all over the world.
technorati tags: IBM, systems, performance, Oracle, benchmark,SAP,SD,SD-Parallel,data,distribution,Sun,benchmarking