Welcome to MilkyWay@home

Posts by Freeze_XJ

1) Message boards : Number crunching : Linux Has Massively Superior AMD OpenCL Performance to Windows (Message 60271)
Posted 1 Nov 2013 by Freeze_XJ
Post:
Not a facepalm, that's not the Linux way :D

First: get the drivers directly from AMD, in most cases the stable one will be sufficient. (http://support.amd.com/en-us/download/desktop?os=Linux+x86 )

Unpack them in a nice spot (any place will do), and open a commandline (still the easiest way to do things). Navigate to that spot (main commands: ls and cd <directory name> ) and make the file (amd-catalyst<long name>.run) executable with chmod +x <filename>. Then execute it ( sudo ./<filename> ) and click the install buttons. Done? Restart computer (it will ask you to do so), and play with BOINC from there on. Good luck!
2) Message boards : News : Separation Modified Fit v1.24 (Message 59683)
Posted 26 Aug 2013 by Freeze_XJ
Post:
Errors are still occurring, with following in stdout

Failed to calculate integral 0
Failed to calculate likelihood
<background_integral> nan </background_integral>
<stream_integral> nan nan nan </stream_integral>
<background_likelihood> nan </background_likelihood>
<stream_only_likelihood> nan nan nan </stream_only_likelihood>
<search_likelihood> nan </search_likelihood>
16:59:27 (9512): called boinc_finish

(http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=548714650)
Running on an XUbuntu 13.04, AMD7950 (factory-OCed), and nothing out of ordinary.

In the past few days I've seen 11 errors (out of 200+ WUs), all exactly identical.
3) Message boards : News : Nbody 1.04 (Message 56721)
Posted 4 Jan 2013 by Freeze_XJ
Post:
I missed this topic before (and accidentally started my own in Number Crunching), but to me, the 104s are multithreaded, which seems to work well. However runtimes vary between 10 seconds and several hours, I pre-emptively aborted a 62(!)hour WU, since my computer may well shutdown before that time, and I'm not willing to test checkpointing on such a large job.

Although the WU says it's OpenCL, it does nothing with my GPU, has the -np 6 --device 0 options, and is only using CPU. Credits are thus low to very low, but that might be correct.
4) Message boards : Number crunching : Credit and CPU utilisation of NBody_104 (Message 56718)
Posted 4 Jan 2013 by Freeze_XJ
Post:
When I'm looking in my awarded credits for the fresh NBody runs, they are given only CPU time according to the list, however when monitoring them, they do run on the GPU (though lightly), and do not consume much CPU-power (1% or so). As expected the awarded credits are abysmally low, see the example. I'm wondering if this is just a stats bug, or an issue with scheduling.
[edit] After some checking, the 104s DO use a lot of CPU, and no GPU, even though it says OpenCL. They're quite nice multicore though. [/edit]

374707636 291018734 462964 4 Jan 2013 | 10:51:03 UTC 4 Jan 2013 | 11:55:33 UTC Completed and validated 73.66 3.36 213.76 MilkyWay@Home v1.02 (opencl_amd_ati)
374706893 291018208 462964 4 Jan 2013 | 10:53:21 UTC 4 Jan 2013 | 11:59:49 UTC Completed and validated 224.07 224.07 2.04 MilkyWay@Home N-Body Simulation v1.04 (opencl_amd_ati)
(numbers after 'validated' are <total time> <CPU time> <awarded credit> )

ps: it seems that many many NBodies are either erroring out, or returning different results, so my 'pending' list has grown, although there are a lot of common v1.02s in there.




©2021 Astroinformatics Group