Elisabeth Stahl on Benchmarking and IT Optimization

More Super POWER8: The Long and Short (grain) of IT

with one comment

I know this may sound corny — but I was recently working on a project where I needed to find proverbs. Yes, you read that right, proverbs. Actually several proverbs to introduce several subjects.

At first I thought — this is really contrived, how will this help move the subject along. In fact, this idea really sucks. And then suddenly after spending a few days wading deep into the proverbial proverb sea, I came to realize the deep dark truth: I love proverbs.

Proverbs are short and sweet and speak to our inner feelings better than we could ever express ourselves.

One of my absolutely favorite proverbs is a simple one and I think of it pretty much all the time these days: Talk doesn’t cook rice.

I love this one because it is short and sweet and gets right to the point. Talk does not open the box of rice. Talk does not measure out the rice and the water. And talk certainly does not turn on the stove.

Talk is cheap. Many of us love to talk — at meetings, on the phone, about claims for new products. But is anything really getting cooked ? That’s where the rubber meets the road.

IBM today announced amazing new OpenPOWER based POWER8 systems and a portfolio of outstanding solutions.

Like the scale-out accelerated Linux Power S824L, the scale-up E870 and E880, the IBM Data Engine for Analytics, Power Enterprise Pools for cloud infrastructure. And not just with the talk but with real data, measured out and cooked.

  • As just one example, with the SAP Sales and Distribution benchmark, the new Power E870 POWER8 system was over 2.1x better performance per core than Dell’s x86 system with the brand new “Haswell” chip and over 2.7x better performance per core than the Oracle SPARC M6. (1)
  • See all the new benchmark data in the IBM Power Systems Performance Report. Includes outstanding SPEC CPU and Java results, CPW, and rPerf (now for multiple SMT values !)
  • The new Power E870 and Power E880 Systems support up to 1,000 VMs per system.
  • And the OpenPOWER Foundation now has 59 members – all working together to leverage the IBM POWER processor’s open architecture for broad industry innovation.

You may have seen other new announcements over the past few weeks from a variety of IT providers. But as you read through the claims that have a clear lack of data — you just may find the rice crunchy.


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.

(1)IBM Power Enterprise System E870 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 8 processors / 80 cores / 640 threads, POWER8; 4.19GHz, 2048 GB memory, 79,750 SD benchmark users, running AIX® 7.1 and DB2® 10.5, dialog response: 0.97 seconds, order line items/hour:  8,722,000, dialog steps/hour: 26,166,000, SAPS: 436,100, Database response time (dialog/update): 0.013 sec / 0.026 sec, CPU utilization: 99%, Cert #2014034.  Result valid as of October 3, 2014. Source: vs. Dell PowerEdge R730, on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 2 processors/36 cores/72 threads, Intel Xeon Processor E5-2699v3; 2.30 GHz, 256 GB memory; 16,500 SD benchmark users, running RHEL 7 and SAP ASE 16; Certification # 2014033.vs. Oracle SPARC Server M6-32 on the two-tier SAP SD standard application benchmark running SAP enhancement package 5 for the SAP ERP 6.0 application; 32 processors/384 cores/3072 threads, SPARC M6; 3.60 GHz, 16 TB memory; 140,000 SD benchmark users, running Solaris® 11 and Oracle 11g; Certification # 20014008. Source: Results current as of 10/3/14.

SAP and all SAP logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries. All other product and service names mentioned are the trademarks of their respective companies.

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)

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




Written by benchmarkingblog

October 3, 2014 at 7:47 am

Posted in announcement, POWER8, SAP

Tagged with ,

One Response

Subscribe to comments with RSS.

  1. Hello,
    your conclusion for SAP two-tier benchmark looks ok for the first sight, but on the second man needs to ask: is IBM able to build a machine able to win over M6-32? What I see is that Oracle’s beast is able to process nearly twice SAP users than IBM. Of course, Oracle does have 32 CPUs while IBM *just* 8, but still is IBM able to come up with machine capable of surpassing Oracle’s M6? If so, then any reference to numbers would be great to have!

    BTW: seeing IBM winning on user/per-core ratio is nice, but seeing it being better just 2.7x over M6, this is not so good. POWER8 should be new king of the single-threaded performance. IBM describes it as “8 dispatch, 10 issue, 16 exec pipe” here in HotChips slides: — yet it’s just winning with 2.7x ratio over very modest OoO SPARC design which is only dual-issue? Why is that?



    February 25, 2015 at 12:11 pm

Leave a Reply

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

You are commenting using your 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: