Welcome to MilkyWay@home

Posts by rtennill

1) Message boards : Number crunching : managing windows CPU affinity (Message 65753)
Posted 13 Nov 2016 by rtennill
Post:
Looking for some feedback before going down this rabbit hole...

I can turn off hyperthreading but I like the flexibility it provides. I can leave tasks running while I work or game and there is no noticeable degradation in performance. BOINC tasks will run a bit slower when there's more going on, but I'm ok with that.

I found some powershell examples for getting and setting affinity but I have not tested them out. My thought was to create a scheduled task to look for unconstrained tasks and reassign CPU affinity.

http://stackoverflow.com/questions/19187241/change-affinity-of-process-with-windows-script

Another option is wrapping the command line with a script that automatically generates the 'proper' affinity mask when launching a new task. Basically, generate the mask, write out to a text file for next run, launch boinc task. Next time through, read previous mask, increment as appropriate, etc. I'm not sure if we have access to the command line for each project or not though.

No definitive results just yet but my estimated remaining time skips a second every ~3 seconds. Far from scientific but I just made this change manually on a few tasks so it'll take some time to see how much of an impact it truly has.

Would this be defeating any process/thread management done by BOINC or Milkyway itself?
2) Message boards : Number crunching : HD6970 & Newer versions of WUs (Message 65530)
Posted 24 Oct 2016 by rtennill
Post:
I'm using the latest non-crimson driver with my 6970 and my only issue is getting enough work units to keep it busy.
3) Message boards : Number crunching : AMD Radeon R9 Fury X - app_info.xml and apps - optimizations (Message 65529)
Posted 24 Oct 2016 by rtennill
Post:
Unfortunately this is expected with modern consumer cards.

This project is all about double precision compute capability. The trend for both Nvidia and AMD over the last several years has been to significantly reduce the double precision compute capability on consumer/gaming cards in favor of single precision and other features.

Wikipedia has good tables for comparing the stats across generation and model. My 6970 has more double precision compute than the fury x. My new card, the 7970, will have almost twice the double precision capability.

https://en.wikipedia.org/wiki/AMD_Radeon_Rx_300_series

https://en.wikipedia.org/wiki/Radeon_HD_7000_Series
4) Message boards : News : GPU Issues Mega Thread (Message 65364)
Posted 1 Oct 2016 by rtennill
Post:
Bah I didn't read the other posts here so I missed the announcement that there are no more tasks from the separate mod-fit project....I had disabled that project to run climate CPU tasks a long time ago.

Turned on mw@home proper and started getting tasks....SORRY!!!
5) Message boards : News : Updated Server Daemons and Libraries (Message 65357)
Posted 30 Sep 2016 by rtennill
Post:
Much appreciated. No new work since the 25th and even then I had perhaps a dozen tasks over the course of a week compared to ~ 140k RAC previously.
6) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 63541)
Posted 8 May 2015 by rtennill
Post:
Can't get two WU to run on my ATI 6970 using the configuration posted below. I had set CPU's to 12 but found that MW@H always indicated a full CPU usage and cutting back to 11 while running climateprediction.net cut my MW@H times in half. Any suggestions? Doesn't seem like the app_config.xml is really being used even though it's found since the CPU usage always says 0.949 CPU.

app_config:
<app_config>
<app>
<name>milkyway</name>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.05</cpu_usage>
</gpu_versions>
</app>
</app_config>


account xml:
<resource_share>100</resource_share>
<no_cpu>1</no_cpu>
<no_ati>0</no_ati>
<no_cuda>1</no_cuda>
<project_specific>
<max_gfx_cpu_pct>50</max_gfx_cpu_pct>
<allow_non_preferred_apps>1</allow_non_preferred_apps>
<gpu_target_frequency>60</gpu_target_frequency>
<nbody_graphics_poll_period>30</nbody_graphics_poll_period>
<nbody_graphics_float_speed>5</nbody_graphics_float_speed>
<nbody_graphics_textured_point_size>250</nbody_graphics_textured_point_size>
<nbody_graphics_point_point_size>40</nbody_graphics_point_point_size>
</project_specific>
</project_preferences>
</account>


BOINC log:
CAL: ATI GPU 0: AMD Radeon HD 6900 series (Cayman) (CAL version 1.4.1848, 2048MB, 2016MB available, 6758 GFLOPS peak)
OpenCL: AMD/ATI GPU 0: AMD Radeon HD 6900 series (Cayman) (driver version 1642.5 (VM), device version OpenCL 1.2 AMD-APP (1642.5), 2048MB, 2016MB available, 6758 GFLOPS peak)
OpenCL CPU: Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz (OpenCL driver vendor: Advanced Micro Devices, Inc., driver version 1642.5 (sse2), device version OpenCL 1.2 AMD-APP (1642.5))
Processor: 11 GenuineIntel Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz [Family 6 Model 63 Stepping 2]
Milkyway@Home | Found app_config.xml
Config: simulate 11 CPUs
Config: use all coprocessors




©2024 Astroinformatics Group