SETI orphans
Message boards :
Number crunching :
SETI orphans
Message board moderation
Previous · 1 . . . 29 · 30 · 31 · 32 · 33 · 34 · 35 . . . 43 · Next
| Author | Message |
|---|---|
Richard Haselgrove  Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 
 |
Bah. I thought I was getting somewhere, until...  So, Microsoft, it was you that b****y decided to set up OneDrive for me, you're managing it for me, and now you tell me it's full? Now you know why I stick with Windows 7. 500GB SSD is in the post... |
Ian&Steve C.  Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 
|
FYI, you can run these GPU tasks on openCL 1.1, 1.2 doesnt seem to be absolutely necessary (maybe just an artificial limit project-side) jobs ran fine on my GTX 550ti which is OpenCL 1.1 https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,43352_offset,30#655567 Seti@Home classic workunits: 29,492 CPU time: 134,419 hours  
|
 Raistmer  Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 
 |
Richard, even if you could manage to catch full run it will be hardly usable. Even with 2-3 min log I wait 10-20 seconds on each presentation update... Also, the bigger run, the coarser resolution. Better to start on pause and hit "play" only before one job finishes and hit pause or stop when second job do few gradient steps. If you want to catch the gap between jobs. SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
FYI, you can run these GPU tasks on openCL 1.1, 1.2 doesnt seem to be absolutely necessary (maybe just an artificial limit project-side) I ran it (until driver crash) on GTX 460 SE :) Didn't manage to catch profile log so far though. Only nSight for Visual Studio is able to trace (in theory) OpenCL on NV. So I had to install VS 2019 also. nSight Compute is only for CUDA.... [And of course latest build dropped support of pre-Win10 OSes so had to use older one] SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640
|
I'm going to go even older, I'll see if a GTX 295 will run LOL Seti@Home classic workunits: 29,492 CPU time: 134,419 hours
|
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874
|
I'm going to go even older, I'll see if a GTX 295 will run LOLI've got a 9800 GT we could try... |
|
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640
|
I'm going to go even older, I'll see if a GTX 295 will run LOLI've got a 9800 GT we could try... oh lord LOL Seti@Home classic workunits: 29,492 CPU time: 134,419 hours
|
|
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640
|
I'm going to go even older, I'll see if a GTX 295 will run LOL it took some fiddling but it just picked up a couple OPNG tasks on it. GPU is nearly pegged to 99-100% on both cores (two tasks, dual GPU), i haven't noticed many dips. ~145MB VRAM used temps also fine at about 62C on an open air test bench looks like it'll take an hour to run we'll see if it completes and/or validates, an absolute waste of power though LOL. not bad for an OpenCL 1.0 card (online specs list 1.1, driver reports 1.0) Seti@Home classic workunits: 29,492 CPU time: 134,419 hours
|
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
I'm going to go even older, I'll see if a GTX 295 will run LOL And no driver restarts due to TDR? Or did you disable it already? SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640
|
And no driver restarts due to TDR? Or did you disable it already? I don't know what "TDR" refers to. I didn't really do anything special on the software side. I only edited the coproc_info.xml file to make the card appear to BOINC as OpenCL 1.2 capable (so that WCG would send me a task). the "fiddling" was on the hardware side. the card seems to have broken display outputs. so i used the GTX550ti to run the monitor and disabled the GTX550ti from being used by BOINC with the <ignore_nvidia_dev> flag in cc_config. the card also didnt want to boot or play nice, probably CSM/UEFI issues and conflicts with such a new OS (Ubuntu 20.04) and also could just be some idiosyncrasies with my specific motherboard (ASUS P9X79 E-WS) which is not normal for PCIe layout and has embedded PLX switches for several PCIe slots to expand more lanes. So I played around with different slots until I found a slot that let things boot properly and run. once there, I just installed the driver package as normal (340.108 drivers from Ubuntu graphics drivers PPA). once drivers were installed I noticed no other issues. I did not experience any driver crashes or anything like that. pretty standard stuff. it completed a task fine with no errors, but was marked invalid. The project is having an issue right now with a high number of Invalids though, so I can't say if this is a "real" invalid, or a victim of the current situation. Result Name: OPNG_ 0001892_ 00298_ 1-- Seti@Home classic workunits: 29,492 CPU time: 134,419 hours
|
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874
|
I don't know what "TDR" refers to.TDR is a Windows specific concept, but I think it accounts for a large proportion of the iGPU 'time limit exceeded' errors under Windows. |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
I don't know what "TDR" refers to.TDR is a Windows specific concept, but I think it accounts for a large proportion of the iGPU 'time limit exceeded' errors under Windows. http://developer.download.nvidia.com/NsightVisualStudio/2.2/Documentation/UserGuide/HTML/Content/Timeout_Detection_Recovery.htm SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
If you under Linux you don't need to know about TDR :)) Regarding invalids. As I said before I don't know how they could do validation at all with so different results from run to run on same dataset. SETI apps news We're not gonna fight them. We're gonna transcend them. |
Keith Myers   Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873
|
If you under Linux you don't need to know about TDR :)) My concern also. From looking at my invalids, I wonder just what the criteria or threshold is for the validator. Some tasks have nothing but invalids for all attempts and I surmise a bad task parameter set, but others look like mine are the odd-man out type of invalid where the validated tasks match closely in OS or device or whatever. Sort of what we ran into with same device validations at Seti. Which is not good for producing valid results consistently. Would have to look at the actual result file to see what it contains since the stderr.txt output does not show anything other than successful job completion. Seti@Home classic workunits:20,676 CPU time:74,226 hours   A proud member of the OFA (Old Farts Association) |
|
Raistmer Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121
|
@SETI we at least got exactly same result from run to run on same task. Results could be little different between devices/compilers/OSes, but they were reproducible (at least if app comes from alpha to beta stage of development). Cause SETI data processing is deterministic one. Genetic algorithms for energy of interaction minimization this app uses have random component. It's possible to get reproducible results even in this case (if pseudo-random generator always gives same sequence starting with same initial seed). But... perhaps they don't use same seed or their random number generator more "random" :) EDIT: also, simple error like using value of uninitialized memory cell will give unreproducible results too. Hard to tell w/o diving deeply in code what case is here. SETI apps news We're not gonna fight them. We're gonna transcend them. |
|
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873
|
There are currently a lot invalid tasks being generated or bad results returned. Looks like the worst culprit has been Apple Mac hosts with Nvidia cards. The admin has stopped sending work to those hosts now. Should reduce the current influx of invalid replications being sent out. But back at the start of the non-beta task releases, there were a lot of single replication tasks once the host was validated as a "trusted" host. That seems to have mostly stopped and most of the work I have been getting requires a wingman to validate the task. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
|
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 
|
Got my first GPU tasks on my WCG OpenPandemics-Covid-19. On the BOINC workshop also BOINC@TACC said it was working on Covid-19 but it does not send me any task. Tullio Thet take about 10 minutes on my GTX 1060 using opencl_nvidia_102. |
Joseph Stateson  Send message Joined: 27 May 99 Posts: 309 Credit: 70,759,933 RAC: 3
|
Just realized I got in 300+ WCG GPU tasks I was unable to do an exact count of the valid ones as there was no way to filter on "OPNG" and too many pages to review at their website. 300 Valid -vs- 17 Invalid on 3 different systems, all GTX 1070 or better Unfortunately, my fastest system was not crunching when the boatload of tasks came in nor was it recording BoincTasks history which would have been useful for statistics. |
|
Joseph Stateson Send message Joined: 27 May 99 Posts: 309 Credit: 70,759,933 RAC: 3
|
I don't know what "TDR" refers to.TDR is a Windows specific concept, but I think it accounts for a large proportion of the iGPU 'time limit exceeded' errors under Windows. Thanks, I was unaware of TDR nor that app. Thinking about this, I do not recall seeing "Nvidia driver was reset" warnings recently (like since last year) and I suspect they are being hidden or dealt with another way. I do get occasional and unexpected "desktop refresh" but the event viewer noes not list nvidia which I think is suspicious. nvidia-smi gives the error message "NVidia requested a reboot" on rare occasions which I trap for on my linux temperature reporting app. |
|
Joseph Stateson Send message Joined: 27 May 99 Posts: 309 Credit: 70,759,933 RAC: 3
|
To late to edit my first post but I did manage to find some statistics on completion of the COVID apps Significantly faster times for GPU calculations as shown in attached pic. About 3 minutes to do a calculation as opposed to 2.5 to 6 hours for the CPUs. Not all results were in my Boinctasks history. WCG site shows 300+ results of which 17 were invalid.  ====and best of all====  |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.