BM Seer Unofficial thoughts from an anonymous Sun employee

SPEComp2001 Sun Blade X6440 x86 world Record and measured watts

Thursday Nov 20, 2008

The Sun Blade X6440 server module (2.7 GHz Opteron 8384 'Shanghai') running OpenSolaris 2008.05 and using Sun Studio Express 11/08 compilers delivers a World Record SPECompM2001 x86 16-threads result of 35896.

Sun Blade X6440 server module (2.7 GHz Opteron 8384 'Shanghai') was 19% faster on SPECompM2001 while consuming 9% less power (measured in watts), over the same Sun Blade with AMD Opteron 8356 processors.

very editorial note: Sun continues to measure watts on actual performance benchmark runs on real-sized memory configurations, this number of benchmark expands beyond Sun's CMT platforms this list continues to expand.  The SPEComp  results was run with 32GB on the Sun Blade. If your vendor does not measure watts, then you really need to ask yourself why they avoid publicly showing everyone easy to measure data.

The Sun system beat the Supermicro H8QM8 result by 4%, even though the X6440 uses DDR667 memory and the Supermicro uses DDR800 memory. This was made possible by using OpenSolaris 2008.5 plus the Sun Studio Express 11/08 compiler rather than SuSE Linux 10 SP1 with the PathScale compiler used by Supermicro.

Press Release content on www.sun.com

SPEComp2001 Performance Chart - SPECompM2001 (bigger is better, ordered by peak)

Result Cores Chips Thrds System
Peak Base
35896 32843 16 4 16 Sun Blade X6440 (Opteron 8384 2.7GHz)
34415 33086 16 4 16 Supermicro H8QM8 (Opteron 8384 2.7GHz)
30275 29094 16 4 16 AMD Tyan Thunder 425QE (Opteron 8360 2.5GHz)
30228 27568 16 4 16 Sun Blade X6440 (Opteron 8356 2.3GHz)
28283 27001 16 4 16 AMD Tyan Thunder 425QE (Opteron 8356 2.3GHz)

Benchmark Description

The SPEC OMPM2001 Benchmark Suite was released in June 2001 and tests HPC performance using OpenMP for parallelism.

  • 11 programs (3 in C and 8 in Fortran) parallelized using OpenMP API
Goals of suite:
  • Targeted to mid-range (4-32 processor) parallel systems
  • Run rules, tools and reporting similar to SPEC CPU2006
  • Programs representative of HPC and Scientific Applications

Power data in watts all results using OpenSolaris 2008.05, watts are measured on a single blade.  As with any vendor multiple blades can be put into a single chassis.

Configuration System State Max. Power drawn in watts
Single X6440 Blade 32GB (16x2GB DDR667) 4x Opteron 8384 2.7GHz During run of SPECompM2001 532 Watts
Single X6440 Blade 32GB (16x2GB DDR667) 4xAMD Opteron 8356 2.3GHz During run of SPECompM2001 578 Watts
Systemwide Power saving AMD Opteron 8384 vs AMD Opteron 8356 During run of SPECompM2001 46 Watts or 9%
Single X6440 Blade 32GB (16x2GB DDR667) 4x Opteron 8384 2.7GHz System active idle 195 Watts
Single X6440 Blade 32GB (16x2GB DDR667) 4x Opteron 8356 2.3GHz System active idle 205 Watts

Disclosure Statement:

SPEC, SPEComp reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of Nov 19, 2008 and this report. Sun Blade X6440 (4 cores, 4 chips, 16 threads, Opteron 8384 2.7GHz), 35,896 SPECompM2001; Supermicro H8QM8 (4 cores, 4 chips, 16 threads, Opteron 8384 2.7GHz), 34415 SPECompM2001; AMD Tyan Thunder 425QE (4 cores, 4 chips, 16 threads, Opteron 8360 2.5GHz), 30275 SPECompM2001; Sun Blade X6440 (4 cores, 4 chips, 16 threads, Opteron 8356 2.3GHz), 30228 SPECompM2001; AMD Tyan Thunder 425QE0 (4 cores, 4 chips, 16 threads, Opteron 8356 2.3GHz), 28283 SPECompM2001.

Result

X6440 16-threads:
35896 SPECompM2001
Reference Date:
Nov 18, 2008
System:
Sun Blade X6440
Total Number Processors:
4
Total Memory :
32 GB (16x2GB DDR2-667MHz)
Processor/GHz of Server:
AMD Opteron 8384, 2.7 GHz
Operating System:
OpenSolaris 2008.05
Compiler:
Sun Studio Express 11/08

[5] Comments
Like this post? del.icio.us | furl | slashdot | technorati | digg
Posted at 07:39AM Nov 20, 2008 in Benchmarks
Tags:
Comments:

"very editorial note: Sun continues to measure watts on actual performance benchmark runs on real-sized memory configurations" - like the MySQL consolidation on the T5220 where the power calculator was used?-) I meant to ask about that config because I couldn't find the mix of options on a 1.2 GHz T5220 to get to your stated 570 watts there, but I digress...

With what sort of meter was power measured on this blade? An in-chassis sort of thing?

Posted by rick jones on November 20, 2008 at 08:28 AM PST #

These are MEASURED WATTS with very accurate meters. When are you ever going to ask HP or IBM or DELL or anyone else for their measurement???
I search their blogs and can't find you commenting on them? Why?

Watts on Sun benchmarks are always measured. Sun would compare against other published results, buy other vendors avoid them like the plague. why? why? ok we know why...

Sun also measures watts on many competitive systems and validates with power calculators from those vendors. For public comparison Sun points to that other vendor's power calculators then decrease those watts to give them the benefit of the doubt.

The MySQL consolidation you mention was measured on both Sun and Dell and validated by the customer.

Why aren't you screaming about the LV-DIMMS used only in SPECpower benchmarks and not where else? SPEC and SPECpower are reg trademarks of Standard Performance Evaluation Corporation, for more info see www.spec.org.

Posted by BM Seer on November 20, 2008 at 09:21 AM PST #

http://blogs.sun.com/bmseer/entry/mysql_consolidation_on_sun_sparc

has been updated to show what was measured and why it references power calculators. Sorry for the confusion.

Posted by BM Seer on November 20, 2008 at 09:34 AM PST #

Rick you seem to keep a close look at Sun's data, so I imagine you look at other vendors. Can you please share interesting watt results from other blogs and publications you've seen?

Posted by BM Seer on November 20, 2008 at 09:44 AM PST #

Assuming the X6440 is like the X6240:

http://www.sun.com/servers/blades/ doesn't seem to show the X6440

where was the meter attached? Was it measuring the entire chassis with just the one blade in it? The entry text "watts are measured on a single blade" suggests it was some sort of intra-chassis power meter.

As for those other vendors, clearly it must be that I find your blog the most entertaining :) More seriously though, I'm not quite sure where to find those other blogs anyway. Feel free to point them out and I can take a look.

WRT the MySQL consolidation, I'm still confused the by mix of measured and power calculator watts. The power calculator for the T5220 at 1.2 GHz with 32GB (16x2) of RAM and 2x146 disc (what _does_ N2 mean there vs T2 anyway?) lists 461 Watts for the 100% busy SPECjbb2005 workload. Either that workload used in the power calculator is light on power on the T5220 vs MySQL, or there were more "things' in the T5220 than listed in the blog entry.

The reason I'm curious about the number of other "things" (discs, PCIe cards, etc) in the T5220 is the 1.2 GHz, 32GB, 2x146 GB disc configuration comes-out at $24495 based on what I see for Config 2 at http://shop.sun.com/is-bin/INTERSHOP.enfinity/WFS/Sun_NorthAmerica-Sun_Store_US-Site/en_US/-/USD/ViewStandardCatalog-Browse?CategoryName=SPARC_T5220&CategoryDomainName=Sun_NorthAmerica-Sun_Store_US-SunCatalog. That makes the list price for 24 of them $587880 and if I divide that by the annual difference in cost for power ($115059) it makes the break-even point ~5.11 years.

But the big difference between the stated measured power and the power calculator suggests there was even more in each T5220, which means each would have a higher list price, which would push the break-even point out even further. Before discounts of course.

And that is without getting into the power cost figures being based on the systems all going flat-out 24x7x365.

One other nit/typo about that entry - the table says 700 instances but the text later-on says 751.

Posted by rick jones on November 20, 2008 at 07:09 PM PST #

Post a Comment:
Comments are closed for this entry.