Message boards :
Number crunching :
Setting up Linux to crunch CUDA90 and above for Windows users
Message board moderation
Previous · 1 . . . 129 · 130 · 131 · 132 · 133 · 134 · 135 . . . 162 · Next
Author | Message |
---|---|
Joseph Stateson Send message Joined: 27 May 99 Posts: 309 Credit: 70,759,933 RAC: 3 |
I also have GEN=4 (two identical socket 1050 TB85) and I had to disable the on board graphics on the system with the I7 chip. The other board, the one I just posted about) had a Xeon chip which did not have intel graphics. I originaly tried getting the intel gen4 hd graphics to run OpenCL but intel does not support Gen4 or earlier from what I read so the 3nd board I built I got a xeon which was a lot cheaper. You might check your on-board graphics and see if the desktop is there or simply make sure the onboard is disabled on that gigabyte HTH |
elec999 Send message Joined: 24 Nov 02 Posts: 375 Credit: 416,969,548 RAC: 141 |
I have it disabled. I can see the computer post, and ubuntu boot. Then it goes to a black screen. |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
I would say an update bunged up the Nvidia drivers. Try reverting to Nouveau drivers in recovery mode and the reinstall the Nvidia drivers. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
elec999 Send message Joined: 24 Nov 02 Posts: 375 Credit: 416,969,548 RAC: 141 |
Anyone know what this error means? <core_client_version>7.14.2</core_client_version> <![CDATA[ <message> process exited with code 1 (0x1, -255)</message> <stderr_txt> Cuda error 'Couldn't get cuda device count ' in file 'cuda/cudaAcceleration.cu' in line 154 : no CUDA-capable device is detected. </stderr_txt> ]]> |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
It means that the SETI application couldn't find any CUDA devices when it tried to run a SETI task. The SETI (CUDA) application wouldn't have been launched by BOINC if BOINC couldn't find any capable GPUs. So the cards, or drivers, must have been working when BOINC started, but must have failed subsequently. Something must have changed. What were you doing on the machine after starting BOINC, but before getting that message? |
elec999 Send message Joined: 24 Nov 02 Posts: 375 Credit: 416,969,548 RAC: 141 |
It means that the SETI application couldn't find any CUDA devices when it tried to run a SETI task. I didnt do anything, this is a dedicated machine. I noticed its giving these errors after the gpu is done processing the work. |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
It means that the SETI application couldn't find any CUDA devices when it tried to run a SETI task. Besides the obvious GPU malfunction, there is a high probability of something corrupts your host. . Normally a simple reset of the host solves the problem. |
rob smith Send message Joined: 7 Mar 03 Posts: 22190 Credit: 416,307,556 RAC: 380 |
Or a GPU has failed. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
elec999 Send message Joined: 24 Nov 02 Posts: 375 Credit: 416,969,548 RAC: 141 |
Does CPU,ram have a huge effect on results. I noticed my dual 2060 doing really well. Beating even my super card and the 2070. |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Since you seem to be referring to gpu performance, the gpus need enough cpu support to run their best. Since data has to flow through both the memory and cpu to get from the gpu task in storage to the gpu card to crunch, if either or both of those components are sped up, they will increase the throughput of all gpus in the host. [Edit]Looking at those hosts I also see a difference in BOINC version, OS versions and kernel versions. Generally the 5.0 kernels are seeing a speed regression from the earlier kernels like 4.15.0. For reference look at some of the charts and tests at phoronix.com for kernel benchmark testing. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 |
Does CPU,ram have a huge effect on results. I noticed my dual 2060 doing really well. Beating even my super card and the 2070. For GPU work, no. Just make sure the CPU isn’t overcommitted (ie trying to run 100% CPU work + GPU work on top) and you’ll be fine. I run one of my systems with low power 2.4GHz CPUs and DDR3 ram with zero impact to the GPU work. Seti@Home classic workunits: 29,492 CPU time: 134,419 hours |
Keith Myers Send message Joined: 29 Apr 01 Posts: 13164 Credit: 1,160,866,277 RAC: 1,873 |
Does CPU,ram have a huge effect on results. I noticed my dual 2060 doing really well. Beating even my super card and the 2070. YMMV. I do see a difference in system performance comparing hosts with slower cpu and memory clocks compared to faster cpu and memories clocks on the same cpu and hardware on other hosts. Seti@Home classic workunits:20,676 CPU time:74,226 hours A proud member of the OFA (Old Farts Association) |
TBar Send message Joined: 22 May 99 Posts: 5204 Credit: 840,779,836 RAC: 2,768 |
Does CPU,ram have a huge effect on results. I noticed my dual 2060 doing really well. Beating even my super card and the 2070. If you compare the CPU tasks of those two systems you will see the CPU with the 2070 is Over-committed compared to the one with the two 2060s. GPUs Must receive free CPU time to run their best. It's the Number One Item in the ReadME. Try reducing the CPU tasks on the machine with the 2070 using the Use at most __ % of the CPUs setting. 2070 CPU = Computer ID 8765031 Run time 2 hours 16 min 39 sec CPU time 1 hours 43 min 27 sec 2060 CPU = Computer ID 8775954 Run time 1 hours 45 min 14 sec CPU time 1 hours 44 min 41 sec |
elec999 Send message Joined: 24 Nov 02 Posts: 375 Credit: 416,969,548 RAC: 141 |
Is there anyway of knowing without checking the boinc client if both gpus are being used? |
Sleepy Send message Joined: 21 May 99 Posts: 219 Credit: 98,947,784 RAC: 28,360 |
Is there anyway of knowing without checking the boinc client if both gpus are being used?nvidia-smi? |
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 |
Is there anyway of knowing without checking the boinc client if both gpus are being used? Please be more specific when asking questions. You have 16 active hosts, and each one with different OS/hardware configs. Without context no one knows what you are asking about. include which host you're referencing with your questions. providing a link to the host details page, or at the very least describe which host you are asking about will go a long way. Instead of having the first responses being requests for more information from you. but in general, there are several ways to tell if a GPU is being used. check the nvidia-smi output (linux) or GPUz for windows check the boincmgr 'Tasks' tab and see if you see 2 GPUs processing tasks there put your hand on the GPU, does it feel warm? look at the GPU, are the fans spinning? I looked through all your dual GPU systems, and you have Valid submissions from both GPUs in all cases, so yes, both of your GPUs are being used. Perhaps you can also include WHY you think you are not using them both that you need to ask? Seti@Home classic workunits: 29,492 CPU time: 134,419 hours |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
but in general, there are several ways to tell if a GPU is being use.Look at the std_err text in the results listed on this website. The single host I checked showed tasks being completed on Device 1, and other tasks being completed on Device 2. But I'm not going to check all of them without guidance. |
Ian&Steve C. Send message Joined: 28 Sep 99 Posts: 4267 Credit: 1,282,604,591 RAC: 6,640 |
but in general, there are several ways to tell if a GPU is being use.Look at the std_err text in the results listed on this website. The single host I checked showed tasks being completed on Device 1, and other tasks being completed on Device 2. But I'm not going to check all of them without guidance. yeah, I just checked all of his dual GPU hosts and edited my comment. he has valid work from both GPUs in all cases, so i'm not sure why he is worried that he's not using them. Seti@Home classic workunits: 29,492 CPU time: 134,419 hours |
elec999 Send message Joined: 24 Nov 02 Posts: 375 Credit: 416,969,548 RAC: 141 |
Is there anyway of knowing without checking the boinc client if both gpus are being used? ID: 8759418 Details | Tasks Cross-project stats: BOINCstats.com Free-DC z440 home 46,645.77 2,395,820 7.9.3 GenuineIntel Intel(R) Xeon(R) CPU E5-1620 v3 @ 3.50GHz [Family 6 Model 63 Stepping 2] (8 processors) [2] NVIDIA Quadro P2000 (4095MB) driver: 410.10 Linux Ubuntu Ubuntu 18.04.2 LTS [4.15.0-20-generic|libc 2.27 (Ubuntu GLIBC 2.27-3ubuntu1)] 4 Sep 2019, 14:22:23 UTC I installed a 1050ti this morning. Thats why I asked. I forgot to check before I left the datacenter. |
Kissagogo27 Send message Joined: 6 Nov 99 Posts: 715 Credit: 8,032,827 RAC: 62 |
some examples of validated tasks
|
©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.