| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: OpenPandemics - COVID-19 Project
Thread: OpenPandemics - GPU Stress Test |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 781
|
|
| Author |
|
|
spRocket
Senior Cruncher Joined: Mar 25, 2020 Post Count: 277 Status: Offline Project Badges: 
|
I'm able to grasp enough CPU work but I'm continuously running out of GPU workunits. When I fetch for work I usually get 0, 1 or 2 GPU workunits, not enough to keep my GPU busy (I complete them in +/- 2 minutes). I don't know - I've been getting 16-20 WUs when BOINC decides to report completed work. Could this depend on Intel/Nvidia/AMD? In any case, the stuck uploads don't seem to be affecting my ability to fetch work, just to return it. The returns eventually make it, but it's kind of amusing to see one half of an OPNG unit upload fine and the other half get stuck. I'm wondering if OPNG units need to be made a bit bigger? Mine generally take 3-5 minutes on my GTX 960. |
||
|
|
Buro87 [Lombardia]
Cruncher ITALY Joined: Mar 11, 2014 Post Count: 28 Status: Offline Project Badges: 
|
They will go back to their original state, that is, around 2,000 work units will be sent every 30 minutes. I hope they will provide more wus. Between "normal mode" ->2000wu every30min -> 4.000 wu /h And "stress test mode" -> 8000wu every minute -> 480.000 wu/h possibilitirs are wide. Why not 1000wu every minute for a pace of 60.000 wu/h ?  |
||
|
|
Ian-n-Steve C.
Senior Cruncher United States Joined: May 15, 2020 Post Count: 180 Status: Offline Project Badges: 
|
yeah I'm fairly sure GUI lag is due to the GPU.
----------------------------------------all of my systems experience no additional lag caused by GPU crunching on OPNG. they're all fairly powerful. when I was testing with some older GPUs last week (GTX550Ti) the screen/GUI lag was incredible. weak GPUs are likely more affected by this. either upgrade the GPU, or do as Falcon said, have GPU processing suspend while the computer is in use.  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 |
||
|
|
biini
Senior Cruncher Finland Joined: Jan 25, 2007 Post Count: 334 Status: Offline Project Badges: 
|
What I would need is a script for Windows10. Ask and You shall receive! I had a same problem, so made a little script. ----WARNING---- This script will retry ALL your uploads and downloads, no matter if they are active, pending or stalled, so don't schedule this to run too often. -----WARNING---- use at your own risk: @echo off Also, increase the max_file_xfers_per_project in cc_config.xml |
||
|
|
Ian-n-Steve C.
Senior Cruncher United States Joined: May 15, 2020 Post Count: 180 Status: Offline Project Badges:
|
I'm able to grasp enough CPU work but I'm continuously running out of GPU workunits. When I fetch for work I usually get 0, 1 or 2 GPU workunits, not enough to keep my GPU busy (I complete them in +/- 2 minutes). I don't know - I've been getting 16-20 WUs when BOINC decides to report completed work. Could this depend on Intel/Nvidia/AMD? In any case, the stuck uploads don't seem to be affecting my ability to fetch work, just to return it. The returns eventually make it, but it's kind of amusing to see one half of an OPNG unit upload fine and the other half get stuck. I'm wondering if OPNG units need to be made a bit bigger? Mine generally take 3-5 minutes on my GTX 960. bigger tasks will certainly help the situation. bigger tasks = longer runtime = less frequent work requests = less server load. 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 |
||
|
|
Richard Haselgrove
Senior Cruncher United Kingdom Joined: Feb 19, 2021 Post Count: 360 Status: Offline Project Badges:
|
bigger tasks will certainly help the situation. I can confirm that. I've just collected another trial batch of 100 tasks (dual GPUs) - they were all of the longer-running 13,000 and up batches. I was able to upload and report enough files to be completely clear of uploads before the next top-up was requested.bigger tasks = longer runtime = less frequent work requests = less server load. |
||
|
|
bozz4science
Advanced Cruncher Germany Joined: May 3, 2020 Post Count: 104 Status: Offline Project Badges: 
|
Many of us voiced concerns relating to short runtimes and SSD write volume. And many have suggested to increase runtimes of the WUs. Hopefully this will be discussed thoroughly with the researchers. Runtime increases of just a few minutes up to 4-6 min instead the 2-3 min mark would essentially already cut down server load by half ....
---------------------------------------- AMD Ryzen 3700X @ 4.0 GHz / GTX1660S Intel i5-4278U CPU @ 2.60GHz [Edit 1 times, last edit by bozz4science at Apr 27, 2021 12:57:50 PM] |
||
|
|
Buro87 [Lombardia]
Cruncher ITALY Joined: Mar 11, 2014 Post Count: 28 Status: Offline Project Badges:
|
There is a solution to reduce GUI lag during OPNG crunching? I don't know but that sounds like something that is more related to your GPU. I can even play games running OPN on my RX 550 but I had massive GUI lag while running older GPUs on other projects. Maybe simply set GPU computing to when the system is idle to avoid this problem. I have a old gtx750ti. I don't understand why some projects (like OPNG, Amicable Numbers, Asteroids) produce this heavy GUI lag, while others (like Primegrid, SRBase, Milkyway) they produce only some frame loss during Youtube playback... |
||
|
|
Ian-n-Steve C.
Senior Cruncher United States Joined: May 15, 2020 Post Count: 180 Status: Offline Project Badges:
|
and actually, bigger tasks is what they're pumping out with these high batch number tasks. anything 13,000+ it seems.
----------------------------------------they're running up to 10x longer than the 7000 and under batches on my 2080ti. (13,000+ = ~9 minutes, 7000- = ~1min or less) none have validated yet (pending), the question is, will they receive 10x credit as the older tasks? my gut feeling says no. uplinger has mentioned in a previous posts that the WUs have some maximum credit reward baked in, and that actual credit granted is based on the percentage of calculations needed to get the result. NOT the total calculations. and both old and new tasks are tagged with the same estimated flops size (incorrectly) at 31,450 GFlops, so my guess is that the maximum credit is the same as well. 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 [Edit 1 times, last edit by Ian-n-Steve C. at Apr 27, 2021 1:17:42 PM] |
||
|
|
Biscotto
Cruncher Italy Joined: Apr 11, 2020 Post Count: 27 Status: Offline Project Badges: 
|
yeah I'm fairly sure GUI lag is due to the GPU. All of my systems experience no additional lag caused by GPU crunching on OPNG. they're all fairly powerful. when I was testing with some older GPUs last week (GTX550Ti) the screen/GUI lag was incredible. weak GPUs are likely more affected by this. either upgrade the GPU, or do as Falcon said, have GPU processing suspend while the computer is in use. don't know but that sounds like something that is more related to your GPU. I can even play games running OPN on my RX 550 but I had massive GUI lag while running older GPUs on other projects. Maybe it depends on the gpu and not so much on its age: i had a GTX960 that suffered massive lag when crunching but now i have a RX560 (less powerfull) and can even game when crunching. Something tells me it has to do with "compute mode" on the cards: on the RX560 it sets the gpu freq to the maximum but caps the memory freq; i can't tell for nvidia anymore. It would be interesting if someone dug deeper in this. Papa Ryzen 5 3600 / Mama Radeon RX 560 |
||
|
|
|