Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Thread Type: Sticky Thread
Total posts in this thread: 288
Posts: 288   Pages: 29   [ 1 2 3 4 5 6 7 8 9 10 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 105384 times and has 287 replies Next Thread
caitilarkin
Former World Community Grid Admin
USA
Joined: Nov 4, 2015
Post Count: 331
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
GPU Work Units - Post Your Tech Support Questions Here

Hi everyone,

Please post all tech support questions about GPU work units in this thread. This will help us find and answer concerns more easily. It will also help volunteers who are looking for answers to common questions.

You can learn more about GPU on World Community Grid here.


Thanks for your support!
[Apr 12, 2021 4:39:17 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Falconet
Master Cruncher
Portugal
Joined: Mar 9, 2009
Post Count: 3265
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

I think it could be useful to sticky this very thread on the OPN forum as I feel a significant portion of volunteers will simply go there for their GPU-related queries.
----------------------------------------


AMD Ryzen 5 1600AF 4C/8T 3.2 GHz - 85W
AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W
Intel Z3740 4C/4T 1.8 GHz - 6W
----------------------------------------
[Edit 1 times, last edit by Falconet at Apr 12, 2021 5:59:53 PM]
[Apr 12, 2021 5:57:39 PM]   Link   Report threatening or abusive post: please login first  Go to top 
caitilarkin
Former World Community Grid Admin
USA
Joined: Nov 4, 2015
Post Count: 331
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

Great suggestion, Falconet! I just posted a sticky notice in the OPN forum to direct volunteers to this thread.
[Apr 12, 2021 6:17:49 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Grumpy Swede
Master Cruncher
Svíþjóð
Joined: Apr 10, 2020
Post Count: 1881
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

@Uplinger

Is it correct that the batches we see now contains some tougher cookies to chew on? I'm seeing longer crunching times than on the earlier production WU's.
----------------------------------------

[Apr 12, 2021 9:12:08 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Mumak
Senior Cruncher
Joined: Dec 7, 2012
Post Count: 477
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

The link in OPN forum is not correct, it points to another thread in OPN forum and not here: https://www.worldcommunitygrid.org/forums/wcg...d_thread,43311_offset,190
Or was it perhaps not meant to point here?
----------------------------------------

----------------------------------------
[Edit 2 times, last edit by Mumak at Apr 12, 2021 9:18:10 PM]
[Apr 12, 2021 9:13:07 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Mumak
Senior Cruncher
Joined: Dec 7, 2012
Post Count: 477
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

Posting here again my issue with inability to get GPU units on Intel DG1.

I'm trying to catch a task on this GPU (Intel DG1):
OpenCL: Intel GPU 0: Intel(R) Iris(R) Xe MAX Graphics (driver version 27.20.100.9168, device version OpenCL 3.0 NEO, 6456MB, 6456MB available, 1190 GFLOPS peak)


All settings seem to be OK, I have other machines that are getting WUs, but haven't seen a single GPU task on the Xe yet.
Requesting new tasks for Intel GPU
Scheduler request completed: got 0 new tasks

It doesn't say that No tasks are available for Intel GPU, so I'm starting to wonder whether there's perhaps some limit/filter on the scheduler side.

Here some more details from the coproc_info:
<intel_gpu_opencl>
<name>Intel(R) Iris(R) Xe MAX Graphics</name>
<vendor>Intel(R) Corporation</vendor>
<vendor_id>32902</vendor_id>
<available>1</available>
<half_fp_config>63</half_fp_config>
<single_fp_config>63</single_fp_config>
<double_fp_config>0</double_fp_config>
<endian_little>1</endian_little>
<execution_capabilities>1</execution_capabilities>
<extensions>cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_intel_subgroups cl_intel_required_subgroup_size cl_intel_subgroups_short cl_khr_spir cl_intel_accelerator cl_intel_driver_diagnostics cl_khr_priority_hints cl_khr_throttle_hints cl_khr_create_command_queue cl_intel_subgroups_char cl_intel_subgroups_long cl_khr_il_program cl_intel_mem_force_host_memory cl_khr_subgroup_extended_types cl_khr_subgroup_non_uniform_vote cl_khr_subgroup_ballot cl_khr_subgroup_non_uniform_arithmetic cl_khr_subgroup_shuffle cl_khr_subgroup_shuffle_relative cl_khr_subgroup_clustered_reduce cl_intel_spirv_media_block_io cl_intel_spirv_subgroups cl_khr_spirv_no_integer_wrap_decoration cl_intel_unified_shared_memory_preview cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_intel_planar_yuv cl_intel_packed_yuv cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_3d_image_writes cl_intel_media_block_io cl_khr_gl_sharing cl_khr_gl_depth_images cl_khr_gl_event cl_khr_gl_msaa_sharing cl_intel_dx9_media_sharing cl_khr_dx9_media_sharing cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_intel_d3d11_nv12_media_sharing cl_intel_unified_sharing cl_intel_subgroup_local_block_io cl_intel_simultaneous_sharing </extensions>
<global_mem_size>6769606656</global_mem_size>
<local_mem_size>65536</local_mem_size>
<max_clock_frequency>1550</max_clock_frequency>
<max_compute_units>96</max_compute_units>
<nv_compute_capability_major>0</nv_compute_capability_major>
<nv_compute_capability_minor>0</nv_compute_capability_minor>
<amd_simd_per_compute_unit>0</amd_simd_per_compute_unit>
<amd_simd_width>0</amd_simd_width>
<amd_simd_instruction_width>0</amd_simd_instruction_width>
<opencl_platform_version>OpenCL 3.0 </opencl_platform_version>
<opencl_device_version>OpenCL 3.0 NEO </opencl_device_version>
<opencl_driver_version>27.20.100.9168</opencl_driver_version>
<device_num>0</device_num>
<peak_flops>1190400000000.000000</peak_flops>
<opencl_available_ram>6769606656.000000</opencl_available_ram>
<opencl_device_index>0</opencl_device_index>
<warn_bad_cuda>0</warn_bad_cuda>
</intel_gpu_opencl>

----------------------------------------

----------------------------------------
[Edit 1 times, last edit by Mumak at Apr 12, 2021 9:19:28 PM]
[Apr 12, 2021 9:15:42 PM]   Link   Report threatening or abusive post: please login first  Go to top 
caitilarkin
Former World Community Grid Admin
USA
Joined: Nov 4, 2015
Post Count: 331
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

The link in OPN forum is not correct, it points to another thread in OPN forum and not here: https://www.worldcommunitygrid.org/forums/wcg...d_thread,43311_offset,190
Or was it perhaps not meant to point here?


My mistake! biggrin The link in the OPN forum post has been fixed. Thanks for the correction.
[Apr 12, 2021 9:28:30 PM]   Link   Report threatening or abusive post: please login first  Go to top 
BladeD
Ace Cruncher
USA
Joined: Nov 17, 2004
Post Count: 28976
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

An old thread that was started before this one existed...

GPU WU completes 100% but continues to run
----------------------------------------
----------------------------------------
[Edit 1 times, last edit by BladeD at Apr 12, 2021 9:38:29 PM]
[Apr 12, 2021 9:36:45 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Ian-n-Steve C.
Senior Cruncher
United States
Joined: May 15, 2020
Post Count: 180
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: GPU Work Units - Post Your Tech Support Questions Here

@Uplinger

Is it correct that the batches we see now contains some tougher cookies to chew on? I'm seeing longer crunching times than on the earlier production WU's.


I'd say in general they have more cookies to chew on, but not tougher. more sub-tasks being packaged per WU.

my 2080ti's are still running about the same speed when scaled for the number of sub-tasks. right around 2.5 seconds per sub-task. all systems reporting similar behavior. more jobs in a WU = longer run time = more credit rewarded.
----------------------------------------

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
[Apr 12, 2021 10:02:32 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 288   Pages: 29   [ 1 2 3 4 5 6 7 8 9 10 | Next Page ]
[ Jump to Last Post ]
Post new Thread