Advanced search

Message boards : Graphics cards (GPUs) : GTX 10x0 Under utilised in GPU and RAM

Author Message
Riaan
Send message
Joined: 16 Dec 10
Posts: 4
Credit: 19,812,500
RAC: 0
Level
Pro
Scientific publications
wat
Message 46014 - Posted: 31 Dec 2016 | 4:34:47 UTC

Good day,

I do not know to whom I should address this but I have noticed that GPUGRID under utilise my GTX 1070 by 10% on the GPU and by 50% on the RAM.

I place an extract from GPU-Z below:
Memory Used [MB] , GPU Load [%] , Memory Controller Load [%]
4063 , 90 , 45

Is there anyway of improving this on my side or is it something that has to be done Server side?

I would like it to be as time efficient as possible - the GPU runs at 60C and completes WUs in and around the 12Hr mark.

Regards
Riaan
____________

Betting Slip
Send message
Joined: 5 Jan 09
Posts: 584
Credit: 2,009,270,900
RAC: 1,673,461
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 46018 - Posted: 31 Dec 2016 | 10:36:40 UTC - in response to Message 46014.

Memory is OK
GPU Load is OK You could try using Environment Variable Swan_Sync 1

PappaLitto
Send message
Joined: 21 Mar 16
Posts: 263
Credit: 1,208,145,781
RAC: 4,821,218
Level
Met
Scientific publications
watwat
Message 46019 - Posted: 31 Dec 2016 | 12:16:32 UTC

You're fine and most WUs are fine but one of my cards, GK104 on short WU GERARD_NONAGGRE_cmpd1 is only using 54% utilization and my gpu is at 45C. It's also very long for a short WU.

Profile Bikermatt
Send message
Joined: 8 Apr 10
Posts: 36
Credit: 1,807,103,604
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 46022 - Posted: 31 Dec 2016 | 14:24:47 UTC

You could try running 2 tasks at once with a app_config.xml file in you projects directory. The latest GERARD_MO_ tasks were only giving me 54% GPU utilization on GTX980s in a win7 box with swan_sync enabled. Running two at a time I am getting 80% load. Same story with the GERARD_NONAGGRE tasks on my GTX670s.

<app_config>
<app>
<name>acemdbeta</name>
<gpu_versions>
<gpu_usage>1</gpu_usage>
<cpu_usage>1.00</cpu_usage>
</gpu_versions>
</app>
<app>
<name>acemdlong</name>
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>1.00</cpu_usage>
</gpu_versions>
</app>
<app>
<name>acemdshort</name>
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>1.00</cpu_usage>
</gpu_versions>
</app>
</app_config>

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,828,159,964
RAC: 453,599
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 46023 - Posted: 31 Dec 2016 | 15:12:50 UTC - in response to Message 46022.
Last modified: 31 Dec 2016 | 15:13:39 UTC

There is high GPU utilization with the BNBS tasks (98% on Linux) and they are less CPU restrained, reasonably high utilization with the short SDOEER_CASP22S tasks (91% on Linux/W10), but some of the others are closer to 80% and suffer more performance loss when the CPU is being used elsewhere. If you are getting significantly less than 80% then I suggest you try reducing your CPU usage as that's likely interfering with GPU performance. GL
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Post to thread

Message boards : Graphics cards (GPUs) : GTX 10x0 Under utilised in GPU and RAM