Welcome to MilkyWay@home

Posts by FruehwF

1) Message boards : Number crunching : 7970 crunching more WUs (Message 53814)
Posted 27 Mar 2012 by FruehwF
Post:
Pleasure

[German]
gern gscheng, wir ösi helfen ja gerne mal aus :-)

Ich schreib dir noch ein PN
[/German]
2) Message boards : Number crunching : 7970 crunching more WUs (Message 53723)
Posted 20 Mar 2012 by FruehwF
Post:
Put an app_info.xml File into the Project-Dir of MW@H

Content:

<app_info>
<app>
<name>milkyway</name>
</app>
<file_info>
<name>milkyway_separation_1.02_windows_intelx86__opencl_amd_ati.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>102</version_num>
<flops>1.0e11</flops>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>1</max_ncpus>
<plan_class>ati14ati</plan_class>
<coproc>
<type>ATI</type>
<count>0.5</count>
</coproc>
<cmdline>--gpu-target-frequency 60 --gpu-disable-checkpointing </cmdline>
<file_ref>
<file_name>milkyway_separation_1.02_windows_intelx86__opencl_amd_ati.exe</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>


If the name of the exe file is differen for 64 Windows, you have to replace the name.
Please send back all WU's before you implement the App_info.xml.
then request new WU's
Otherwise they get lost till.

regards

Franz
3) Message boards : Number crunching : No ATI work (Message 53643)
Posted 13 Mar 2012 by FruehwF
Post:
13.03.2012 12:25:03 | Milkyway@Home | No work available
And no Validation of WU's since 10:30 UTC
Obviously I'm not the only one.

what's going on?
4) Message boards : News : New Separation Runs (Message 53556)
Posted 6 Mar 2012 by FruehwF
Post:
Your Catalyst ist to new for the 4850. Downgrade to 11.9!

The newer driver doesn't incluedd something that the 4850 need to work.

See alos:

http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2457&nowrap=true#52751

regardes Franz
5) Message boards : Number crunching : N-Body and the Bunker (Message 53339)
Posted 20 Feb 2012 by FruehwF
Post:
I was able to rebuild Popsie'sproblem, and I think it's realy an issue!

He wants to do only N-body on his machine.
At the start of boinc-manager the machine gets 12 WUs. Every Minute it pulls new WU's, but obviosly there is a limit off maximum 4 WU's per pull But his machine does more then 4 WU's in a minute.
So the work-chache reduces more and more. At the end it pulls only one WU per pull.

The questions are: Why ist there a limit for CPU WU of 4 per request?
Why are the N-Body WU's so damned small?
And finaly: From sight of MW Project-Managers: Does it make sense to do only N-Body?

Deutsch

Ich konnte das Problem von Popsi nachstellen, und ich denke es ist wirklich eine Anforderung:

Er möchte nur N-Body auf seiner Maschine laufen lassen.
Zu Beginn erhält man 12 WU's. Jede Minute erfolgt dann ein automatischer Request für neue WU's. Offensichtlich gibt es aber ein Limt, dass mann nur 4 WU/Request bekommt. Sein Rechner schaft aber mehr als 4 WU's pro Minute, und so wird der Vorrat immer kleiner. Zum Schluss wird gar nur mehr eine WU pro/Request geliefert.
Die Fragen, die sich stellen sind:
Warum gibt es dieses 4 WU Limti pro Request?
Warum sind die N-Body WU's so verdammt klein?
Und schliesslich: Aus Sicht der Projekt-Verantwortlichen: Macht es überhaupt Sinn nur N-Body rechnen zu lassen?
6) Message boards : Number crunching : N-Body and the Bunker (Message 53194)
Posted 15 Feb 2012 by FruehwF
Post:
The WU-Limit for CPU is 3/ Core so you get maximum 12 WU's.
This is a Limt from MW Server i dont think the BM 7 can overrule this.
In the Moment the N-Body WU's are damed short.
IF you only do N-Body you musst do every minute an update (you can do this with boinccmd.exe

[german]

Das WU Limt von MW@H ist 3 / Core, damit bekommst du nur maximal 12 WU's.
Ich denke BM 7 kann das auch nciht übersteueren. Und im Moment sind die N-Body WU's verdammt kurz bei der Laufzeit, wenn man nur N-Body rechnet, dann müsste man jede Minute einen manuellen Update machen (könnte man mit boinccmd.exe automatisieren)

ps: lass deine Sachen nicht durch einen Übersetzer durch, die sind Schrott, schicks vorher lieber mir
[/german]
7) Message boards : Number crunching : 79XX Dont Run (Message 52544)
Posted 16 Jan 2012 by FruehwF
Post:
But in Afterburner you can't downclock the RAM by much.. as it would be efficient for MW. I even asked them some time ago to somehow let people adjust the lower clock limit, to make their tool more useful. They said, it wouldn't be in the spirit of Afterburner..

MrS


Have activated the unofficial overclocking mode? With that I come down on my 4850 till freezing screen.
look here http://forums.guru3d.com/showthread.php?t=338906

regardes Franz
8) Message boards : Number crunching : app_info.xml needed for GPU cuda (Message 51850)
Posted 6 Dec 2011 by FruehwF
Post:
It's not pointless!

The MW WU's do some CPU Work at the end. You can mask this time when you are doing 2 WU's parallel (with a little time shift).

During one WU is doing the end Work on CPU, the other WU runn's 100 % on the GPU. -> You get better Performance .
But Note if you are crunching on CPU also leave one Core/GPU free for MW.

I do my work only on ATI's (they are bette in DP then Nvidias)so I have no app_info.xml, but search on the forum.
regards

Franz


[edit]
I've found something:

- <app_info>
- <app>
<name>milkyway</name>
</app>
- <file_info>
<name>milkyway_separation_0.82_windows_x86_64__cuda_opencl.exe</name>
<executable />
</file_info>
- <app_version>
<app_name>milkyway</app_name>
<version_num>82</version_num>
<flops>1.0e11</flops>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>1</max_ncpus>
<plan_class>cuda</plan_class>
- <coproc>
<type>CUDA</type>
<count>0.5</count>
</coproc>
<cmdline />
- <file_ref>
<file_name>milkyway_separation_0.82_windows_x86_64__cuda_opencl.exe</file_name>
<main_program />
</file_ref>
</app_version>
</app_info>


The <count>0.5</count>
Tag was changed from 1 to 0.5 wich makes 2 WU's run
In the <name> & <file_name> Tag you have to insert your stock_app (32 or 64 bit)

HTH
9) Message boards : Number crunching : 100 Mhz difference (Message 51485)
Posted 24 Oct 2011 by FruehwF
Post:
I did as many did here, who have MW as main project.
I increased one of my 4850's the GPU Clock from 625 up to 750 (20%)and this leads to an increase of performance of 18%.
The memory clock I decreased from 993 Mhz to 490 MHZ, which has no influance to the performance. (saves energy and aviod heat)

In the dual GPU rack I could only go up to 700 MHZ GPU clock because of air-flow Problems. (The amount of invalid WU's raises)
hth
franz
10) Message boards : News : feel free to cancel any in progress WUs (Message 51342)
Posted 10 Oct 2011 by FruehwF
Post:
...

good point.
Collatz gives you the choice to select 'Collatz' or 'Mini Collatz'.
But with larger wu's a checkpointing must be implemented.


Since 0.82 checkpointing is implemented.
11) Message boards : News : feel free to cancel any in progress WUs (Message 51335)
Posted 10 Oct 2011 by FruehwF
Post:
Well I don't know anything about the topologie of the MW - WU so it is just a quess. If an increase of the WU count is not possible. Maybe it's possible to increase the Size of the WU's.
Maybe it's possible to launche an extra application (only for GPU's) with large WU's. So that 1 WU last 1 or 2 hours on a HD5870?
12) Message boards : Number crunching : 34 pages with results (Message 51248)
Posted 28 Sep 2011 by FruehwF
Post:
Yes they have look here:

http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2598&nowrap=true#51146
13) Message boards : Number crunching : separation_assimilator milkyway Not Running !! (Message 51234)
Posted 27 Sep 2011 by FruehwF
Post:
WU's are not valideted, but the clients get new WU's to chruch.

greetings Franz

EDIT: 16:33 UTC, UP and runnig :-) thxs
14) Message boards : Number crunching : GPU Requirements (Message 51161)
Posted 21 Sep 2011 by FruehwF
Post:
seems that ther is a mistake:

wiki:
http://en.wikipedia.org/wiki/Comparison_of_AMD_graphics_processing_units#FirePro_3D_series
FirePro 3D V5800[96] April 26, 2010 Juniper XT(RV840)
But the V5900 has a Cayman proc. so this must have DP
15) Message boards : Number crunching : Why are finished tasks deleted so fast? (Message 51160)
Posted 21 Sep 2011 by FruehwF
Post:
Good thing! :-)

But I think 12 hours are enough to detect faults etc.

With a fast GPU you get tons of WU's in your lists.
A much better improvement will be if we try to increase the WU limit per GPU in steps from 12 to 24 and finaly to 48.
That would bridge small server outages.

regards

Franz
16) Message boards : Number crunching : Feeder not running again :( (Message 51062)
Posted 15 Sep 2011 by FruehwF
Post:
No it's just the same as on tuesday.
Power is off fo a quater of an hour, but MW doesn't start automatic.
So we have to wait till Mark or Travis wake up.
17) Message boards : Number crunching : Feeder not running again :( (Message 51056)
Posted 15 Sep 2011 by FruehwF
Post:
well seems to be indicator how long Matt and Travis are sleeping ;-)
18) Message boards : Number crunching : Computer Lab Alerts & Downtime (Message 51032)
Posted 13 Sep 2011 by FruehwF
Post:
There is a page from cs rpi where the planed downtimes can be look at:

http://www.cs.rpi.edu/twiki/view/LabstaffWeb/LabstaffDowntime
The next downtim is on Thursday 89/15/2011 planned Electrical shutdown 7am-7:15am )
Only 15 Minutes,biut it seeems that Mw doesn't come back automatic.
Can somebody from the stuff can give the milkyway servers a kick to avoid longer outage time as necessary.
19) Message boards : Number crunching : Feeder not running (Message 51021)
Posted 12 Sep 2011 by FruehwF
Post:
Fine!

I don't want to warm up old diskussions, but more cache would be nice in such situations... (24 or 48 WU's per GPU would be a cautious step.)
20) Message boards : Number crunching : Feeder not running (Message 51019)
Posted 12 Sep 2011 by FruehwF
Post:
9/12/2011 planned Electrical shutdown 7am-7:15am
- 15 Minutes, But after that somebody has to kick th machines I think!

Ist there anybody out there?


Next 20

©2019 Astroinformatics Group