Posts by Westsail and *Pyxey*

21) Message boards : Number crunching : GPU issue on new HP311 mini (Message 1060692)
Posted 29 Dec 2010 by Profile Westsail and *Pyxey*
Post:
Someone more knowledgeable can chime in...
but, I don't believe it is possible for to crunch with "shared" memory type video adapter. Something to do with cuda driver implementation or some such I believe..
22) Message boards : Number crunching : ATI MB rev170 Now 177 beta app testing (Message 1060362)
Posted 28 Dec 2010 by Profile Westsail and *Pyxey*
Post:
Some GPU tasks validating now...All looks 5x5 *thumbs up*
hostid=5687772


Got my app info worked out for crunch on cpu+gpu running 4 concurrent MB tasks on dual core CPU + single 4850. They all seem to run around 90-100 minutes. Now maybe i'll get brave and try to add AP in there as well. :)

<app_info>
<app>
<name>setiathome_enhanced</name>
</app>
<file_info>
<name>MB_6.10_win_SSE3_ATI_r177.exe</name>
<executable/>
</file_info>
<file_info>
<name>MultiBeam_Kernels.cl</name>
<executable/>
</file_info>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>610</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>0.05</max_ncpus>
<plan_class>ati13ati</plan_class>
<cmdline>-period_iterations_num 2 -instances_per_device 2</cmdline>
<flops>20987654321</flops>
<file_ref>
<file_name>MB_6.10_win_SSE3_ATI_r177.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>MultiBeam_Kernels.cl</file_name>
<copy_file/>
</file_ref>
<coproc>
<type>ATI</type>
<count>.5</count>
</coproc>
</app_version>
<app>
<name>setiathome_enhanced</name>
</app>
<file_info>
<name>AK_v8b_win_x64_SSE41.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>603</version_num>
<platform>windows_intelx86</platform>
<file_ref>
<file_name>AK_v8b_win_x64_SSE41.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>603</version_num>
<platform>windows_x86_64</platform>
<file_ref>
<file_name>AK_v8b_win_x64_SSE41.exe</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>


Keep crunching & happy new years
-Brandon

edit:
LOL, just realized task views (whatever) is turned off so can't see them anymore... *shrugs*

23) Message boards : Number crunching : ATI MB rev170 Now 177 beta app testing (Message 1060102)
Posted 27 Dec 2010 by Profile Westsail and *Pyxey*
Post:
Running on this HOST now....Intel Core 2 Duo E7500 /w Gigabyyte HD4850 1G.
We'll see once we can report..
Just finished the first 2 running 2x at once... 1 hour 29 minutes & 1 hour 41 minutes.
Not too bad! Still can't believe they are running hopefully they will validate. I upped the worker thread priority and the next set looks to be maybe a tad faster.
So far so good! *thumbs up*
-Brandon
24) Message boards : Number crunching : download problem (Message 1058408)
Posted 21 Dec 2010 by Profile Westsail and *Pyxey*
Post:

Downloads are pegged for goin on a bit now...hope a switch doesn't melt down..

Boinc should sort itself out in time. Just let her cruise..
:)
25) Message boards : Technical News : Maxed (Dec 16 2010) (Message 1057917)
Posted 19 Dec 2010 by Profile Westsail and *Pyxey*
Post:
I don't understand all this bickering about who should and should not get work and suggesting new plans to implement a new "fairness doctorine".

The only completely fair distribution method is the current one. The client makes a request and then either it is successful for not. If not try again. Any other plan is actually unfair to someone.

If you feel that you are being restricted in some way in getting through to Berkeley servers I would suggest you try a different ISP.

+100000
26) Message boards : Number crunching : Limiting users (Message 1057907)
Posted 19 Dec 2010 by Profile Westsail and *Pyxey*
Post:
To add to what was said...Taken just now from my modest GPU host:
SETI@home Enhanced (anonymous platform, nvidia GPU)
Number of tasks completed 5171
Max tasks per day 1290
Number of tasks today 476
Consecutive valid tasks 1191
Average processing rate 137.26068194655
Average turnaround time 1.52 days

That is just for ONE application...Many of these guys run lots faster rigs and multiple applications.

What I can't figure is why a person with a rac of 3 will complain the loudest about being out of work....*scratches head*
27) Message boards : Number crunching : Panic Mode On (41) Server problems (Message 1055870)
Posted 14 Dec 2010 by Profile Westsail and *Pyxey*
Post:
;)
Just sayin'...
We are going at an amazing clip...bandwidth still maxed..Love It!
Nice to be back! =)
28) Message boards : Number crunching : Panic Mode On (41) Server problems (Message 1055866)
Posted 14 Dec 2010 by Profile Westsail and *Pyxey*
Post:


29) Message boards : Number crunching : nVidia Geforce 400 series: How good? (Message 1054820)
Posted 11 Dec 2010 by Profile Westsail and *Pyxey*
Post:
690-720 secs for an original 260-192 here...
1600-1875 looks about right for the 460 3x

Over on Collatz /w the Fermi app 460's are exactly three times faster than 260..So I believe there is hope. :)
30) Message boards : Number crunching : Is the GTX 295 still the best card for CUDA? (Better than a GTX 470?) (Message 1050242)
Posted 21 Nov 2010 by Profile Westsail and *Pyxey*
Post:
Absolutely thrilled with Fermi here...Over in Collatz the 460 is exactly 3x as fast as my 260 and 10 degrees cooler. :)
<10 minutes for 460 vs. 30+ minutes on the old 260
Haven't had a chance to compare on Seti yet..Sounds like when Jason gets done they will rock'n roll!

Jason, Will there be multiple apps for different versions of GF1xx or a single app for all Fermi? I ask cause of the architecture differences between 460/580 and 465/470/480. Thanks in advance!
31) Message boards : Number crunching : Welcome to the 11 year club (Message 1050091)
Posted 21 Nov 2010 by Profile Westsail and *Pyxey*
Post:
Better late than never ehh..
Gratz everybody! Can't believe there are so many folks that have been here through the whole ride. I think that speaks volumes...
Still remember my first host was a pentium 100 or so mhz (133?) IBM Aptiva
Built an AMD K6-2 machine o/c (ie. 266mhz>289mhz or some such) around '99 for crunching. Oh the memeories...
SETI@home classic workunits 1,503
SETI@home classic CPU time 25,425 hours

Cheers! Hope to see us all still around for the 20th anniversary..
32) Message boards : Cafe SETI : U.S. building next wave of supercomputers (Message 1048914)
Posted 17 Nov 2010 by Profile Westsail and *Pyxey*
Post:
BOINC as a whole is averaging 5.1 Petaflops, this was averaged in April though.

WOW, very interesting..Thanks! First time i've seen a quantitative measure..

From what I read, all new supercomputers use GPUs as coprocessors. SETI@home may have been a pioneer in this trend.

Yea, I wondered if that was what they meant by
system that will include accelerators.
*scratches head* Accelerators? Like in Switzerland? lol

The last one I remember reading about was a Japanese computer TSUBAME. It uses Nvidia GPU :)

Tokyo Tech Builds First Tesla GPU Based Heterogeneous Cluster To Reach Top 500
33) Message boards : Cafe SETI : U.S. building next wave of supercomputers (Message 1048806)
Posted 16 Nov 2010 by Profile Westsail and *Pyxey*
Post:
Two 20-petaflop systems planned for 2012, but China's Tianhe-1A expected to rank No.1 on Top500 list this year

Computerworld - The U.S. is building two 20-petaflop supercomputers, many times more powerful than anything operating today, including China's new supercomputer, the Tianhe-1A, which is expected to be officially crowned next week as the world's fastest system.

Oak Ridge National Laboratory, home for what has been the world's most powerful system, the Jaguar, a 1.75-petaflop system, versus Tianhe-1A's 2.5 petaflops, is building a 20-petaflop system that will include accelerators.

That system will be ready in 2012, James Hack, director of the National Center for Computational Sciences at Oak Ridge, told Computerworld. No other details about the system are being offered.

Another 20-petaflop system is being built for the Lawrence Livermore National Laboratory by IBM. That system has already been announced and is expected to arrive at the lab in late 2011 and be in production in 2012.
The earliest it is likely to be ready for consideration in the Top500 Supercomputer Sites list will be for the June 2012 release of the ranking, said Don Johnston, a spokesman for Lawrence Livermore labs.

Whether these 20-petaflop systems emerge as the top systems in the world remains to be seen, but with China also racing ahead in building its own systems, supercomputing is becoming intensely globally competitive.

"Personally I love it," Jeremy Smith, director of director of the Center for Molecular Biophysics at Oak Ridge, said of the international attention now being paid to supercomputing. "In competing with other countries everybody gains and wins - that's why I'm excited about it."

The global attention may raise the profile of supercomputing and help keep government interest in funding it. When Japan's Earth Simulator emerged as the world's fastest computer in 2002, it "shocked the supercomputing world," Smith said.

The National Research Council, in a report soon after the Japan achieved the top ranking, said that Japan's system "has served as a wake-up call, reminding us that complacency can cause us to lose not only our competitive advantage but also, and more importantly, the national competence that we need to achieve our own goals."

But the Japanese system was a one-time occurrence. China, in contrast, has embarked on a sustained drive to not only build world leading systems, but its own processors and interconnecting technologies for high-performance systems as well.

Japan is still building supercomputers, but its major disclosed project is a 10-petaflop system, nicknamed the "K computer." It's being developed by Fujitsu for use by High-Performance Computing Infrastructure Initiative by Japan's Ministry of Education, Culture, Sports, Science and Technology, and is due in 2012.

cont at:LINK

So I wonder how BOINC as whole ranks on that list?? It should be included too..An international virtual volunteer based supercomputer. ;)
34) Message boards : Number crunching : Panic Mode On (41) Server problems (Message 1046144)
Posted 5 Nov 2010 by Profile Westsail and *Pyxey*
Post:
I'm only in it for all the toasters I can buy with the credits.


And all this time, I thought it was the other way round...

least the house is warm tonight :)
35) Message boards : Number crunching : Name the newest member to the server family!!! (Message 1045901)
Posted 3 Nov 2010 by Profile Westsail and *Pyxey*
Post:
Another vote for Scotty here!
36) Message boards : Number crunching : 32b vs 64b (Message 937926)
Posted 6 Oct 2009 by Profile Westsail and *Pyxey*
Post:
Very interesting guys!

Is there any difference, at the present time, between running 32b vs 64b on a GPU host?

Had 64b machines for awhile now but still run all 32, as I haven't seen the need to switch. Will get win7 soon...if..I can run the stuff we need, would like to go win764 on all our xp32 machines. hmmm.. *scratches head*
37) Message boards : Number crunching : - POLL - : SETI@home on ATI GPU.. (Message 937922)
Posted 6 Oct 2009 by Profile Westsail and *Pyxey*
Post:
Yes, yes and yes.. ;)
38) Message boards : Number crunching : Heat Alarm (Message 931836)
Posted 7 Sep 2009 by Profile Westsail and *Pyxey*
Post:

In my opinion, we're going to find a dead fan.


Exactly. I would shut it down, if mine..Until I could confirm fans etc all working and give it a good cleaning. Then can start to figure out baseline/max temps.

My experience is after say 6 months, generally any machine could at least benefit 'some' from cleaning of any dust/contamination etc.
39) Message boards : Number crunching : Heat Alarm (Message 931826)
Posted 7 Sep 2009 by Profile Westsail and *Pyxey*
Post:
I recommend CPUID HW Monitor

Maybe post a screenshot for us if you like. This should display the output from all your sensors and has logging and high/low values.
Good luck. Heat is the biggest killer of PC hardware.
Very nice studio btw!
Keep crunching
40) Message boards : Number crunching : Replica DB is back (Message 931542)
Posted 7 Sep 2009 by Profile Westsail and *Pyxey*
Post:
WOW! *applause* way to go guys! Echoing what someone said above..New hardware rocks..


Previous 20 · Next 20


 
©2026 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.