Welcome to MilkyWay@home

Posts by DJStarfox

21) Message boards : Number crunching : Need help to use less cores (Message 50916)
Posted 1 Sep 2011 by DJStarfox
Post:
OK good. I'm sure you found my thread on this topic. :)
http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2301

And this one too:
http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2411
22) Message boards : Number crunching : N-Body + Mac OS X + i7 = no good ? (Message 49406)
Posted 18 Jun 2011 by DJStarfox
Post:
They just released n-body 0.60 for mac osx.
23) Message boards : News : Nbody updated to 0.60 (Message 49405)
Posted 18 Jun 2011 by DJStarfox
Post:
Can linux users just stick with nbody 0.40? You just did this previous version in March.
24) Message boards : Number crunching : Is there a way to limit the number of cpus... (Message 48835)
Posted 17 May 2011 by DJStarfox
Post:
Elliotts,
Just go with the same solution that Matt posted. Limit MW tasks to at most 4 processors on your i7 (odd numbers are also valid). I've set BOINC & MW to only use 4 CPUs, and MW tasks finish in 8 minutes.
25) Message boards : Number crunching : AMD and NV GPU working together? (Message 47579)
Posted 11 Apr 2011 by DJStarfox
Post:
How to make a cc_config.xml file:
http://boinc.berkeley.edu/wiki/Client_configuration
26) Message boards : Number crunching : AMD and NV GPU working together? (Message 47128)
Posted 7 Apr 2011 by DJStarfox
Post:
You probably need the following options inside the cc_config.xml (under <options>):

<ignore_ati_dev>0</ignore_ati_dev>
<ignore_cuda_dev>1</ignore_cuda_dev>
<use_all_gpus>1</use_all_gpus>
27) Message boards : Number crunching : Multi-threaded nBody (Message 46945)
Posted 3 Apr 2011 by DJStarfox
Post:
To set the nBody application to only use a specific number of processors (or just 1 CPU), you have to create an app_info.xml file like the one at the top of this thread. Just replace the avg_ncpus, max_ncpus, and nthreads values with 1 instead of 4.

However, the project should do this CPU limiting automatically.
28) Message boards : News : N-body updated to 0.40 (Message 46930)
Posted 2 Apr 2011 by DJStarfox
Post:
Every single one of my nbody units seem to be erroring out instantly. Can't tell if its a problem on my end or MW's :(


This is the error you're getting:
<core_client_version>6.10.58</core_client_version>
<![CDATA[
<message>
process exited with code 22 (0x16, -234)
</message>
<stderr_txt>
execv: No such file or directory

</stderr_txt>
]]>


I would check/reset file system permissions!
29) Message boards : News : N-body updated to 0.40 (Message 46886)
Posted 1 Apr 2011 by DJStarfox
Post:
can someone please post a working app_info.xml for the new application?


See my first post:
http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2301
30) Message boards : Number crunching : Multi-threaded nBody (Message 46871)
Posted 31 Mar 2011 by DJStarfox
Post:
Running 4 threads, I had one finish in 27 minutes.
Running 8 threads, I had one finish in 8 minutes.

Seems a bit off.
31) Message boards : Number crunching : Multi-threaded nBody (Message 46870)
Posted 31 Mar 2011 by DJStarfox
Post:
Anyone willing to run some nbody apps with 1, 2, 4, and 8 threads to compare times? I'm curious how much the speedup really is.

If you have an app_info.xml file, add the following param (changing the 4's at the end to whatever):
<app><!-- stock app for N-Body -->
        <name>milkyway_nbody</name>
        <user_friendly_name>MilkyWay@Home nBody</user_friendly_name>
    </app>
    <file_info>
        <name>milkyway_nbody_0.40_x86_64-pc-linux-gnu__mt</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>milkyway_nbody</app_name>
        <version_num>40</version_num>
        <file_ref>
          <file_name>milkyway_nbody_0.40_x86_64-pc-linux-gnu__mt</file_name>
          <main_program/>
        </file_ref>
        <avg_ncpus>4</avg_ncpus>
        <max_ncpus>4</max_ncpus>
        <cmdline>--nthreads=4 </cmdline>
    </app_version>
32) Message boards : News : N-body updated to 0.40 (Message 46868)
Posted 31 Mar 2011 by DJStarfox
Post:
I noticed that the workunits I've been getting run for about 2 seconds before erroring out. Is it a 0.40 issue or a workunit issue?

For some reason there are still some lingering old workunits (though they should have stopped completely, the percentage of them is decreasing), and the new version rejects the old format, so it's a workunit issue.


Funny enough, I tried to keep both versions in my app_info.xml file, so I could process both old and new WU. But BOINC will delete the old version from disk upon startup!

I hope there are still a few people out there running the old version long enough to empty the system of the older WU.
33) Message boards : News : N-body updated to 0.40 (Message 46861)
Posted 31 Mar 2011 by DJStarfox
Post:
BOINC controls the number of threads used. If you want to set the number, you use something like <avg_ncpus> and <max_ncpus> in app_info.xml. You also might be able to control it more directly with the command line control that I think exists in app_info.xml by adding something like <cmdline>--nthreads 4</cmdline>, though I'm not sure how that will play with how BOINC wants to schedule things.


<avg_ncpus> and <max_ncpus> only TELL boinc how many CPUs a sci app uses; it does NOT dictate to the sci app. I tried param --nthreads=4 and set the ncpus params both to 4. Hard-coding the number of threads seems to work. Thanks for the tip.
34) Message boards : News : N-body updated to 0.40 (Message 46858)
Posted 31 Mar 2011 by DJStarfox
Post:
WAIT ONE MINUTE!

How do I set the number of CPUs that the nbody 0.40 app uses inside my app_info.xml file? Each process is using 4 CPUs, but BOINC doesn't seem aware of this.

Edit: Scratch that; each process uses ALL my CPUs without regard to any BOINC preferences about MAX CPUs!

How do I limit how many threads this new nBody app can use ?!?
35) Message boards : News : Updated separation to 0.50 (Message 45798)
Posted 26 Jan 2011 by DJStarfox
Post:
For some reason, if I let BOINC go back to stock, it doesn't use the sse2 app for NBody. So, I'm forcing the i686 SSE2 app with my app_info.xml file.
There aren't any non-SSE2 N-body applications.


I wasn't sure because it's not listed as such on the applications page.
http://milkyway.cs.rpi.edu/milkyway/apps.php

If everything requires SSE2, then simplier to just leave off the "(see2)". In 2011, I'd bet 99.9% of the CPUs attached to this project support SSE2.
36) Message boards : News : Standalone screensaver test available for Linux (Message 45767)
Posted 25 Jan 2011 by DJStarfox
Post:
This is a great start, Shane.

At some point--when you're done, I hope someone can help you convert the SDL graphics calls to OpenGL. That will quadruple the framerate and lower CPU usage for the screensaver app.
37) Message boards : News : Updated separation to 0.50 (Message 45766)
Posted 25 Jan 2011 by DJStarfox
Post:
My current applications are:
milkyway_0.18_sse4.2_x86_64-pc-linux-gnu
milkyway_nbody_0.07_x86_64-pc-linux-gnu

Those are really old. That nbody I don't think should never validate; something changed after that.


OK, I updated to:
milkyway_0.50_x86_64-pc-linux-gnu__sse2
milkyway_nbody_0.21_i686-pc-linux-gnu__sse2

For some reason, if I let BOINC go back to stock, it doesn't use the sse2 app for NBody. So, I'm forcing the i686 SSE2 app with my app_info.xml file.

As long as someone posts a thread such as "Updated MW to version 0.50" in the News forum, my RSS reader will alert me so I can update my applications.
38) Message boards : Application Code Discussion : Code Ignores BOINC Network Settings (Message 45726)
Posted 23 Jan 2011 by DJStarfox
Post:
I think I figured out why I'm observing these messages. The milkyway@home applications do not generate a result file, but instead they put the result in the stderr section of the scheduler request (reporting as complete). There is no result file to upload.
39) Message boards : Application Code Discussion : Code Ignores BOINC Network Settings (Message 45661)
Posted 21 Jan 2011 by DJStarfox
Post:
Maybe, but I can prove that BOINC does this. There is a bug somewhere, even if it's not in the milkyway code specifically.

20-Jan-2011 09:00:00 [---] Suspending network activity - time of day
20-Jan-2011 09:01:56 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:06:57 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:11:56 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:16:57 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:21:57 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:26:58 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:31:58 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:36:58 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:41:58 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:46:58 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:51:59 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 09:56:59 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:01:59 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:07:00 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:11:59 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:17:00 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:22:00 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:27:00 [Milkyway@home] [task_debug] result de_nbody_model6_4_8721_1295496950_0 checkpointed
20-Jan-2011 10:28:48 [Milkyway@home] [task_debug] Process for de_nbody_model6_4_8721_1295496950_0 exited
20-Jan-2011 10:28:48 [Milkyway@home] [task_debug] task_state=EXITED for de_nbody_model6_4_8721_1295496950_0 from handle_exited_app
20-Jan-2011 10:28:48 [Milkyway@home] [task_debug] exit status 0
20-Jan-2011 10:28:48 [Milkyway@home] [task_debug] result state=FILES_UPLOADING for de_nbody_model6_4_8721_1295496950_0 from CS::app_finished
20-Jan-2011 10:28:48 [Milkyway@home] [task_debug] result state=FILES_UPLOADED for de_nbody_model6_4_8721_1295496950_0 from CS::update_results
40) Message boards : Application Code Discussion : Code Ignores BOINC Network Settings (Message 45656)
Posted 20 Jan 2011 by DJStarfox
Post:
The NBody simulation code completely ignores the BOINC preferences for network availability. When I suspend all network activity, the Milkyway app will still try to upload its result when finished. This needs to be fixed.

Should mention I'm using boinc 6.10.45 x86_64. All other projects that I'm attached to respect the network preferences.


Previous 20 · Next 20

©2024 Astroinformatics Group