Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Thread Type: Sticky Thread Total posts in this thread: 290
|
![]() |
Author |
|
hnapel
Advanced Cruncher Netherlands Joined: Nov 17, 2004 Post Count: 82 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
how would I go about making this app configuration file I am good with computers but not great with coding I have this question too.Hi, The exact numbers to put in will depend on your specific hardware. Higher powered GPUs can run more of these opng units simultaneously. To create the file open up a text editor such as notepad++ or the default windows notepad. Put the below code into it, adjust to your hardware, and save it as app_config.xml in the project directory. I think the default on windows is "C:\ProgramData\BOINC\projects\www.worldcommunitygrid.org". Then open up boinc manager and go to options->read local prefs files. If that does not exist, you are in simple view and need to switch to advanced view: View->Advanced view (or press Ctrl+Shift+a). This example is for 2 work units to be processed at the same time, reserving 1 thread per gpu workunit. You can try that and see what the average gpu utilization is and then increase as necessary. To increase replace the .5 in gpu_usage with 1/Number desired. So 3=.33, 4=.25, etc. If you have a more complex setup, like running a discrete gpu and an integrated gpu of a different brand, they can be setup differently. If that is the case, let me know. I hope this helps! ![]() <app_config> I would like my iGPU to have less jobs than my NVIDIA GPUs, how would my app_config.xml look like in that case? |
||
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
how would I go about making this app configuration file I am good with computers but not great with coding I have this question too.Hi, The exact numbers to put in will depend on your specific hardware. Higher powered GPUs can run more of these opng units simultaneously. To create the file open up a text editor such as notepad++ or the default windows notepad. Put the below code into it, adjust to your hardware, and save it as app_config.xml in the project directory. I think the default on windows is "C:\ProgramData\BOINC\projects\www.worldcommunitygrid.org". Then open up boinc manager and go to options->read local prefs files. If that does not exist, you are in simple view and need to switch to advanced view: View->Advanced view (or press Ctrl+Shift+a). This example is for 2 work units to be processed at the same time, reserving 1 thread per gpu workunit. You can try that and see what the average gpu utilization is and then increase as necessary. To increase replace the .5 in gpu_usage with 1/Number desired. So 3=.33, 4=.25, etc. If you have a more complex setup, like running a discrete gpu and an integrated gpu of a different brand, they can be setup differently. If that is the case, let me know. I hope this helps! ![]() <app_config> I would like my iGPU to have less jobs than my NVIDIA GPUs, how would my app_config.xml look like in that case? It is similar to above, but you have to use the plan_class tag. Here is an example app_config.xml set to run 4 nvidia work units per GPU and 1 per intel GPU. It also illustrates reserving more cores per intel GPU work units. I reserve extra because my analysis showed that running CPU + iGPU resulted in less total science done (jobs/hr(gpu work units have far more jobs than CPU work units)) than iGPU without CPU work units. <app_config> If you are running a different gpu, you can look in task manager(windows)/ps or htop, etc(linux) or in the [Boinc Data Directory]->projects->www.worldcommunitygrid.org folder. You are looking for the last part of filenames like this one: wcgrid_opng_autodockgpu_7.28_windows_x86_64__opencl_intel_gpu_102 That would fall under the opencl_intel_gpu_102 plan_class. The 102 means openCL has to be at least 1.2. References: Client_Configuration AppPlan ![]() |
||
|
tech4theworld
Cruncher Joined: May 16, 2011 Post Count: 2 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I have a system where the detection of the GPUs by OpenCL is not as expected. The Log shows the two Nvidia GPUs are detected (correctly) in the CUDA lines, but OpenCL shows two lines per physical GPU. The result is two WUs dispatch to GPU0 and none to GPU1. Does anyone have a suggestion for overrides to correct this so one WU is dispatched to each GPU?
Excerpt from log file: 5/1/2021 9:08:26 AM | | cc_config.xml not found - using defaults 5/1/2021 9:08:27 AM | | Starting BOINC client version 7.16.11 for windows_x86_64 5/1/2021 9:08:27 AM | | log flags: file_xfer, sched_ops, task 5/1/2021 9:08:27 AM | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2s zlib/1.2.8 5/1/2021 9:08:27 AM | | Data directory: C:\ProgramData\BOINC 5/1/2021 9:08:27 AM | | Running under account xxxxxx 5/1/2021 9:08:27 AM | | CUDA: NVIDIA GPU 0: Quadro K620 (driver version 462.31, CUDA version 11.2, compute capability 5.0, 2048MB, 1689MB available, 863 GFLOPS peak) 5/1/2021 9:08:27 AM | | CUDA: NVIDIA GPU 1: Quadro K620 (driver version 462.31, CUDA version 11.2, compute capability 5.0, 2048MB, 1689MB available, 863 GFLOPS peak) 5/1/2021 9:08:27 AM | | OpenCL: NVIDIA GPU 0: Quadro K620 (driver version 462.31, device version OpenCL 1.2 CUDA, 2048MB, 1689MB available, 863 GFLOPS peak) 5/1/2021 9:08:27 AM | | OpenCL: NVIDIA GPU 0: Quadro K620 (driver version 462.31, device version OpenCL 1.2 CUDA, 2048MB, 1689MB available, 863 GFLOPS peak) 5/1/2021 9:08:27 AM | | OpenCL: NVIDIA GPU 1: Quadro K620 (driver version 462.31, device version OpenCL 1.2 CUDA, 2048MB, 1689MB available, 863 GFLOPS peak) 5/1/2021 9:08:27 AM | | OpenCL: NVIDIA GPU 1: Quadro K620 (driver version 462.31, device version OpenCL 1.2 CUDA, 2048MB, 1689MB available, 863 GFLOPS peak) 5/1/2021 9:08:27 AM | | Windows processor group 0: 8 processors Best Regards |
||
|
Keith Myers
Senior Cruncher USA Joined: Apr 6, 2021 Post Count: 193 Status: Offline Project Badges: ![]() |
You have two sets of video drivers installed. Use DDU to remove all the drivers and then reinstall the latest drivers.
----------------------------------------That is why your cards are detected twice. Common problem and standard fix. ![]() A proud member of the OFA (Old Farts Association) |
||
|
johncmacalister2010@gmail.com
Veteran Cruncher Canada Joined: Nov 16, 2010 Post Count: 799 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Can someone please direct me to the list of acceptable GPUs? I cannot find it.....
----------------------------------------Thanks! ![]() crunching, crunching, crunching. AMD Ryzen 5 2600 6-core Processor with Windows 11 64 Pro. AMD Ryzen 7 3700X 8-Core Processor with Windows 11 64 Pro (part time) ![]() |
||
|
Ian-n-Steve C.
Senior Cruncher United States Joined: May 15, 2020 Post Count: 180 Status: Offline Project Badges: ![]() |
Can someone please direct me to the list of acceptable GPUs? I cannot find it..... Thanks! pretty much any GPU that's OpenCL 1.2 capable. depending on OS used, some users have reported issues with older GPUs on Windows, even with compatible OpenCL 1.2, low end GT 700-series cards seem to be a recurring problem child. Linux seems to have less issues. ![]() EPYC 7V12 / [5] RTX A4000 EPYC 7B12 / [5] RTX 3080Ti + [2] RTX 2080Ti EPYC 7B12 / [6] RTX 3070Ti + [2] RTX 3060 [2] EPYC 7642 / [2] RTX 2080Ti |
||
|
Hal Bregg
Cruncher Joined: Nov 20, 2018 Post Count: 5 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Hello,
I noticed that Intel GPU is used and GTX950 sits idle. What would be correct app_config.xml settings to enable Nvidia card? |
||
|
biini
Senior Cruncher Finland Joined: Jan 25, 2007 Post Count: 334 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
<use_all_gpus>1</use_all_gpus>
----------------------------------------in cc_config.xml |
||
|
goben_2003
Advanced Cruncher Joined: Jun 16, 2006 Post Count: 146 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Can someone please direct me to the list of acceptable GPUs? I cannot find it..... Thanks! There is not really a list, it should be (almost?) all cards that support openCL 1.2. If BOINC detects your card, it should be listed along with which openCL it has in the event log when BOINC starts up. ![]() |
||
|
Hal Bregg
Cruncher Joined: Nov 20, 2018 Post Count: 5 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
<use_all_gpus>1</use_all_gpus> in cc_config.xml Thanks. I'll try to snatch some of the WUs now. |
||
|
|
![]() |