GPU on my main cruncher getting slower and slower

Message boards : Number crunching : GPU on my main cruncher getting slower and slower
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Shannon Lester
Avatar

Send message
Joined: 27 Jul 09
Posts: 83
Credit: 12,388,119
RAC: 140
United States
Message 1995992 - Posted: 30 May 2019, 19:23:40 UTC

Everyone helped me a couple months ago get my ASUS Expedition GTX 1060 6gb tuned and running great on my main crunching rig (Dell T7400 Dual Xeon 5450 8 Cores). With everyone's help we had the gpu doing tasks in about 7 minutes. Well.... it has gotten slower and slower, I was seeing 7 minutes, then 8 something, 9 and now today work units are completing around 10:32 so something is up.

I am still running the command line given to me:

-sbs 1024 -period_iterations_num 1 -tt 1500 -high_perf -high_prec_timer -spike_fft_thresh 4096 -tune 1 64 1 4 -oclfft_tune_gr 256 -oclfft_tune_lr 16 -oclfft_tune_wg 256 -oclfft_tune_ls 512 -oclfft_tune_bn 64 -oclfft_tune_cw 64 -hp


and 1 CPU + 1 NVIDIA GPU in app_config.xml, computing preferences are:

Use at most 100 % of the CPUs
Use at most 100 % of CPU time

I'm stumped as to why this is happening...

Also I really need to get my other machine to running better with the help from my VEGA thread but I have been so busy with work that I haven't had time to work on it and I've just left the main cruncher to its own.

All of this has happened before and will happen again -Battlestar Galactica
ID: 1995992 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1995993 - Posted: 30 May 2019, 19:27:02 UTC - in response to Message 1995992.  

How many work units are you running and how many are on the GPU?
ID: 1995993 · Report as offensive
Profile Shannon Lester
Avatar

Send message
Joined: 27 Jul 09
Posts: 83
Credit: 12,388,119
RAC: 140
United States
Message 1995995 - Posted: 30 May 2019, 19:31:04 UTC - in response to Message 1995993.  

1 work unit at a time on the gpu and 7 on the remaining cpu cores.

All of this has happened before and will happen again -Battlestar Galactica
ID: 1995995 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 37564
Credit: 261,360,520
RAC: 489
Australia
Message 1995997 - Posted: 30 May 2019, 20:16:32 UTC

It's just the variation in the workunits available and nothing wrong at your end, if fact my 3750K rig (Win7) has recorded its lowest RAC ever under just Green Bank work.

Cheers.
ID: 1995997 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11449
Credit: 29,581,041
RAC: 66
United States
Message 1996000 - Posted: 30 May 2019, 20:24:20 UTC
Last modified: 30 May 2019, 20:24:36 UTC

To add to Gary's post it is well know Nvidea GPUs don't really do well on VLARs which is what Green Banks data is. IIRC at on time the project wouldn't even send them to Nvideas.
ID: 1996000 · Report as offensive
Ian&Steve C.
Avatar

Send message
Joined: 28 Sep 99
Posts: 4267
Credit: 1,282,604,591
RAC: 6,640
United States
Message 1996018 - Posted: 30 May 2019, 21:48:10 UTC - in response to Message 1995995.  

Reduce your CPU use to 5 CPU jobs. You may be starving out the CPU resources available to properly feed the GPU.
Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

ID: 1996018 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19534
Credit: 40,757,560
RAC: 67
United Kingdom
Message 1996021 - Posted: 30 May 2019, 21:53:35 UTC
Last modified: 30 May 2019, 21:54:13 UTC

On my GPU the tasks take ~250 seconds or ~310 seconds, depending on blc2n, at the moment blc25 are taking 310s and blc26 250s.
ID: 1996021 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1996022 - Posted: 30 May 2019, 21:57:53 UTC

You are currently crunching through the re-introduced BLC25 data. That is some of the hardest and longest running BLC data we have seen in a year. Nothing wrong with your setup. The RAC is just going to fall when Green Bank data is all we get. Accept it.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1996022 · Report as offensive
Profile Shannon Lester
Avatar

Send message
Joined: 27 Jul 09
Posts: 83
Credit: 12,388,119
RAC: 140
United States
Message 1996034 - Posted: 30 May 2019, 23:02:43 UTC - in response to Message 1996022.  

Dang that’s rough! Well I’m glad that there’s nothing wrong because I’ve been racking my brain trying to figure out if something had happened that I wasn’t aware of. I’m still learning too so thanks sir! :)

All of this has happened before and will happen again -Battlestar Galactica
ID: 1996034 · Report as offensive
halfempty
Avatar

Send message
Joined: 2 Jun 99
Posts: 97
Credit: 35,236,901
RAC: 114
United States
Message 1996198 - Posted: 1 Jun 2019, 3:43:41 UTC

Thanks all for posting about this. I was in the same boat racking my brain trying to figure out what I could have unwittingly messed up to make my RAC steadily decrease like this.
ID: 1996198 · Report as offensive

Message boards : Number crunching : GPU on my main cruncher getting slower and slower


 
©2025 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.