BM Seer Unofficial thoughts from an anonymous Sun employee

« Configs used for... | Main | Sun World Record for... »

Sun Fire X4450 SPECjbb World Records

Friday May 09, 2008

Dave Dagastine's Blog covers Sun's two new SPECjbb2005 World Records on Sun's X64 systems. The Sun Fire X4450, 4-chip Xeon QC CPUs and Java SE 6 Update 6-P, now hold the 4-chip Multi-JVM World Record and the single-JVM x86 world record. see: http://blogs.sun.com/dagastine/entry/sun_java_on_intel_delivers1

Now one of the things I don't know is if my Sun colleages used a non-standard BIOS on this benchmark. I imagine my colleages did the same thing as the other vendors who benchmark Xeons on SPECjbb, so at least everyone is playing the same field. My personal opinion is that the BIOS should not be tuned differently for different benchmarks. But no one listens to me

Disclosure Statement

SPEC, SPECjbb reg tm of Standard Performance Evaluation Corporation. Sun Fire X4450 results submitted to SPEC. Other results as of 05/6/08 on www.spec.org. Sun Fire X4450 (4 chips, 16 cores, Sun JDK 6u6-p) SPECjbb2005 bops = 464355, SPECjbb2005 bops/JVM = 58044. Sun Fire X4450 (4 chips, 16 cores, Sun JDK 6u6-p) SPECjbb2005 bops = 464355, SPECjbb2005 bops/JVM = 58044. SPECjbb2005 bops = 389208, SPECjbb2005 bops/JVM = 389208.

[6] Comments

Like this post?

Posted at 05:00AM May 09, 2008 in Benchmark Issues
Tags: performance specjbb

Comments:

What a wonderful juxtaposition. In the blog entry previous to this:

http://blogs.sun.com/bmseer/entry/configs_used_for_specpower_ssj we read:

"It is so easy to measure watts on benchmarks that one publishes, Sun does it all of the time."

And in this one, and the one referenced by it what do we see? Well, we don't see any watts :)

Posted by rick jones on May 09, 2008 at 08:10 AM PDT #

... oh, if I were king!

Actually others inside Sun complained bitterly about this. I think we'll see a lot more soon. Maybe you could complaint to HP, IBM, Dell,... they won't listen to me. Maybe they'll listen to you.

Posted by BM Seer on May 09, 2008 at 10:14 AM PDT #

Ah - are HP, IBM and Dell bloggers also complaining that SPEC doesn't require power on everything while also claiming to publish power figures for all their benchmarks and then not actually doing so? If so, feel free to point me at those blogs and I'll be happy to comment there as well.

Posted by rick jones on May 09, 2008 at 03:51 PM PDT #

It is much faster than 2x more expensive T5240 ;)
T5240: bops=373405, SPECjbb2005 bops/JVM=23338
X4450: bops=464355, SPECjbb2005 bops/JVM=58044

Posted by Triffids on May 10, 2008 at 12:18 AM PDT #

Rick: you really missed my point. I just don't see the new IBM SPECpower report yet. I know it will show up on spec.org soon.

I am not saying HP/IBM/Dell are claiming anything incorrectly. I'm just saying they need to show power on every benchmark configuration as sun is marching toward doing.

Posted by BM Seer on May 12, 2008 at 12:10 PM PDT #

Sun is starting to push out watt/perf on other benchmarks:
see this new one:
http://blogs.sun.com/bmseer/entry/sun_fire_x4440_best_opteron

Or last week's VMmark:
http://blogs.sun.com/bmseer/entry/vmmark_performance_watt_performance

While others vendors talk about only about chip-watts, peak watts, or test small&low-GHz configs, Sun is on the path to show measured power on full performance benchmarks.

Posted by BM Seer on May 15, 2008 at 01:19 PM PDT #