Doing the Right Thing: Another SAP #1 on POWER7
Sometimes it seems as if everything around us is one big DON’T.
- Don’t say bad things about the lives of your ex-fans, like LeBron.
- Don’t tweet pictures of yourself from the gym.
- Don’t forget to lock your door in a hotel room.
- Don’t hike in shorts during black fly season.
- And don’t publish a benchmark — unless you have something cool to say about it.
Which IBM just did this week – a lovely jewel of a present for the IBM 100th birthday. A new #1 SAP SD 2-tier 12-core result on the IBM Power 730 POWER7 system — with Linux SLES 11 and DB2.(1)
This IBM result handily surpassed the HP SLES 11 result on the 12-core DL380.(2)
And for more on doing the right thing see what IBM is working on this week.
************************************************
(1)SAP SD 2-tier result of 5250 SAP SD benchmark users, IBM Power 730, POWER7, 3.7 GHz (2 processors / 12 cores / 48 threads), SLES 11, IBM DB2 9.7, SAP enhancement package 4 for the SAP ERP 6.0 (Unicode), 0.98 seconds average dialog response time, Certification #2011022.
(2)SAP SD 2-tier result of 4757 SAP SD benchmark users, HP DL380 G7, Xeon X5680, 3.33 GHz (2 processors / 12 cores / 24 threads), SLES 11, MaxDB 7.8, SAP enhancement package 4 for the SAP ERP 6.0 (Unicode), 0.98 seconds average dialog response time, Certification #2011001.
Source: http://www.sap.com/benchmark. Results current as of 6/15/11.
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.
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: IBM, systems, performance, POWER7, Xeon,730,Linux,SLES,SAP,gcc,DB2,9.7,HP,DL380,Xeon,X5680,MaxDB,benchmark,benchmarking
This is compare to x86 server, how about the comparision of price and price/performance between P730 vs DL380 G7 ?
Rex
June 23, 2011 at 10:48 am
Hi Rex,
Thanks for your comment. You are so right. Performance is just one aspect of the equation. The SAP SD 2-tier benchmark employs performance as a metric. But you should also consider other requirements such as price performance, availability, security, virtualization technologies, etc.
benchmarkingblog
June 23, 2011 at 11:02 am
$5 x86-machine and open source db lost less than 10% to a half million ibm-machine. are you sure this is cool ?
Triffids
August 18, 2011 at 3:03 pm
Hi Triffids,
Good to hear from you. Even though price isn’t a metric in this benchmark, you still might want to verify the costs you are quoting with your vendor reps. And remember that you also need to consider other requirements besides performance and price performance such as availability, security, virtualization technologies, sophistication of databases, etc. The cool thing here is that you can use this all as input to make your decision.
benchmarkingblog
August 18, 2011 at 4:51 pm
if compare performance why don’t you take this x86 results ?
6846 users IBM System x3690 X5, 2 Processors / 20 Cores
6230 users HP ProLiant BL620c G7, 2 Processors / 20 Cores
Triffids
August 19, 2011 at 3:42 am
Yes, see http://public.dhe.ibm.com/eserver/benchmarks/news/newsblurb_x3690X5_Windows_sap_081611.pdf .
benchmarkingblog
August 19, 2011 at 1:53 pm