OpenCL NV MultiBeam v8 SoG edition for Windows

Message boards : Number crunching : OpenCL NV MultiBeam v8 SoG edition for Windows
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 7 · 8 · 9 · 10 · 11 · 12 · 13 . . . 18 · Next

AuthorMessage
Joe Januzzi
Volunteer tester
Avatar

Send message
Joined: 13 Apr 03
Posts: 54
Credit: 307,134,110
RAC: 492
United States
Message 1778804 - Posted: 13 Apr 2016, 20:10:51 UTC - in response to Message 1778773.  

Zalster,
I didn't touch my app-info, and I got this on my start up log. Looks like I'm going to wait and see, before I try to fix my file.
Joe

4/13/2016 2:49:34 PM | SETI@home | Found app_info.xml; using anonymous platform
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: setiathome_v7 windows_intelx86 700
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: setiathome_v7 windows_x86_64 700
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 703 sse2
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 703 sse2
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 703 sse
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 703 sse
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 701
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 701
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 700
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 700
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 710 opencl_nvidia_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 710 cuda_opencl_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 710 opencl_nvidia_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 710 cuda_opencl_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 705 opencl_nvidia_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 705 cuda_opencl_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 705 opencl_nvidia_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_x86_64 705 cuda_opencl_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 710 opencl_nvidia_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 710 cuda_opencl_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 710 opencl_nvidia_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 710 cuda_opencl_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 705 opencl_nvidia_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 705 cuda_opencl_100
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 705 opencl_nvidia_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: astropulse_v7 windows_intelx86 705 cuda_opencl_cc1
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: setiathome_v8 windows_x86_64 800
4/13/2016 2:49:34 PM | SETI@home | [error] State file error: duplicate app version: setiathome_v8 windows_intelx86 800


Real Join Date:
Joe Januzzi (ID 253343) 29 Sep 1999, 22:30:36 UTC
Try to learn something new everyday.
ID: 1778804 · Report as offensive
Bruce
Volunteer tester

Send message
Joined: 15 Mar 02
Posts: 123
Credit: 124,955,234
RAC: 11
United States
Message 1778821 - Posted: 13 Apr 2016, 21:15:48 UTC

I have never seen this type of error in the log before, they must have made some changes in the server to cause them, probably because of GBT.

You do not need both platforms. If you have a 32bit OS, then stay with the intelx86 platform. If you have a 64bit OS, you can use either one.

I have a Win x64 and I use the intelx86 platform in my app-info.

If you have a 64bit CPU and a 64bit OS, then you have a x86_64 platform.

You can delete the the sections that include your duplicates.

Make backups!

Hope this helps.
Bruce
ID: 1778821 · 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 1778826 - Posted: 13 Apr 2016, 21:38:48 UTC - in response to Message 1778804.  
Last modified: 13 Apr 2016, 21:39:18 UTC

Joe,

The only error I got was the one listed but I removed everything from the setiathome folder and restall the lunatics then cut the last section out and pasted in the SoG section.

Don't know if that made a difference or not.

I do know this, when I looked at the app_info I do have see both x86-64 and intelx86 mixed thru out it, so why it only flagged those 2 is beyond me.

Installed the latest version of Raismter SoG

It seems to show better utilization but I'm only at 3 MB per GPU and none of them have been Guppi yet so don't know it will respond.

@ Bruce

Going to keep an eye on them for a while and I'll debate about removing that 1 section from my app_info

Zalster
ID: 1778826 · Report as offensive
Joe Januzzi
Volunteer tester
Avatar

Send message
Joined: 13 Apr 03
Posts: 54
Credit: 307,134,110
RAC: 492
United States
Message 1778829 - Posted: 13 Apr 2016, 21:56:36 UTC - in response to Message 1778826.  

Zalster,
I loaded my back-up app-info file, and no more errors in my start up log. I saved the corrupt file(?) for CYA. What version of SoG are you running? Where do you keep track on the latest version?
Joe

Real Join Date:
Joe Januzzi (ID 253343) 29 Sep 1999, 22:30:36 UTC
Try to learn something new everyday.
ID: 1778829 · 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 1778860 - Posted: 14 Apr 2016, 0:48:35 UTC - in response to Message 1778829.  

Joe, I got it from Raistmer's dropbox.

https://cloud.mail.ru/public/DMkN/x4BRCYuAV

I'm thinking it's beta 8.12

but I just keep everything as 800 here on main
ID: 1778860 · Report as offensive
Joe Januzzi
Volunteer tester
Avatar

Send message
Joined: 13 Apr 03
Posts: 54
Credit: 307,134,110
RAC: 492
United States
Message 1778865 - Posted: 14 Apr 2016, 1:01:06 UTC - in response to Message 1778860.  

Zalster,
Thanks for the info:-)
Joe

Real Join Date:
Joe Januzzi (ID 253343) 29 Sep 1999, 22:30:36 UTC
Try to learn something new everyday.
ID: 1778865 · 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 1779219 - Posted: 15 Apr 2016, 4:41:41 UTC

Beta has Guppi for testing on GPUs

I'm moved my machine there.

So far with OpenCL_nvidia_sah it's just under 16 minutes for a guppi

Need to clear those out then try the OpenCl_nvidia_SoG on them
ID: 1779219 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22537
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1779225 - Posted: 15 Apr 2016, 5:12:06 UTC

Thanks for the heads-up, I've just suspended main on one of mine and rolled it over to Beta. Like you a pile of "normal" tasks to get through, then let it "run wild" on the next delivery....
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1779225 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22537
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1779227 - Posted: 15 Apr 2016, 5:36:47 UTC

....just got the first few guppi - mixture of SoG, Cuda42 & Cuda50. It will be interesting to see how they compare in terms of run times (should be running in a few hours time, some of the GPU tasks will be on a GTX980 and some on GTX970s (and maybe a few on GTX780s....)
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1779227 · 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 1779233 - Posted: 15 Apr 2016, 5:40:33 UTC
Last modified: 15 Apr 2016, 5:40:51 UTC

Sah (16 mins) is twice as fast as cuda 50 (33 minutes), still waiting for some SoGs
ID: 1779233 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22537
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1779271 - Posted: 15 Apr 2016, 10:27:25 UTC
Last modified: 15 Apr 2016, 10:31:39 UTC

SoG approx the same as SAH, both approx half CUDA50

(CUDA50 time about two-three times "normal" Arecibo MB)

Edit to add:
All on the same GTX980
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1779271 · 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 1779314 - Posted: 15 Apr 2016, 15:10:15 UTC
Last modified: 15 Apr 2016, 15:13:51 UTC

Was observing the SoG and noticed they start off fast but when they hit 70% complete they start to crawl.....dragging out the completion

example...

6 minutes into computations, 70% done then they slow down and take another 6 minutes to finish those last 30%

I'm wondering if this the fix Raistmer came up with to prevent the 100% usage of the CPU when nearing the completion of the work unit.
ID: 1779314 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1779318 - Posted: 15 Apr 2016, 15:27:20 UTC - in response to Message 1779314.  

Which precise SoG version was this, please? If it was for NV card, please cite r number and I'll check the actual <frac_done> values with my modded client - it was a progress reporting problem last time.
ID: 1779318 · 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 1779319 - Posted: 15 Apr 2016, 15:33:29 UTC - in response to Message 1779318.  

I'm using the last one Raistmer had on his dropbox

Mb8_win_x86_SSE3_OpenCl_NV_r3430_SoG.7z
ID: 1779319 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1779323 - Posted: 15 Apr 2016, 15:59:46 UTC - in response to Message 1779319.  
Last modified: 15 Apr 2016, 16:53:14 UTC

OK, got it, thanks. Did you happen to notice any task details - what AR, etc.? Hooked it up to Beta and got guppi_VLAR as the first fetch.

It does seem to be an issue related to the one we were discussing at Beta - but at VLAR, perhaps not so drastic.

wu_name: blc3_2bit_guppi_57451_20612_HIP62472_0007.12525.831.18.21.146.vlar
WU true angle range is :  0.008175

            <prog>        <fraction_done>
17:13:25
17:14:25    0.01748931    0.043302
17:15:25    0.04244621    0.098808
17:16:26    0.06566018    0.159925
17:17:27    0.09050095    0.215225
17:18:32    0.11596886    0.277116
17:19:36    0.14290232    0.341525
17:20:36    0.16774924    0.396144
17:21:43    0.18374324    0.442169
17:22:50    0.20764196    0.497730
17:23:56    0.23405194    0.559920
17:24:59    0.26072407    0.622799
17:26:03    0.28700744    0.684736
17:27:04    0.32917878    0.704677
17:28:05    0.37018072    0.726249
17:29:06    0.41237318    0.745490
17:30:07    0.45388002    0.766894
17:31:08    0.49561945    0.785576
17:32:08    0.53885943    0.807092
17:33:08    0.58042432    0.826390
17:34:10    0.62295983    0.845603
17:35:10    0.66387519    0.866467
17:36:11    0.70599017    0.885643
17:37:12    0.74688920    0.906469

The SoG application keeps two different counters for how far it's got: "Progress" and "Fraction done". We're seeing fraction done in the progress column in BOINC Manager. You'd expect them to be the same, but clearly either fraction done is counting up too fast in the early stages, or progress is counting more slowly (which is what we're used to from the CUDA apps, which - in reporting terms - start slow and speed up). I don't think there's any definite answer for which is right - perhaps we need to arrange another developer brainstorming party, if that desert island is still free.
ID: 1779323 · 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 1779338 - Posted: 15 Apr 2016, 17:08:26 UTC - in response to Message 1779323.  
Last modified: 15 Apr 2016, 17:23:25 UTC

Looking all the SoG I've run this am, they are all 0.008175

I'll look farther back to last night

Edit..

If it is just the progress we see versus the fraction done than I guess there is nothing we can do.

I do like that they don't suddenly jump up to 100% of CPU like they were doing.

Still 13 minutes GPU is much better than 55 minutes CPU
ID: 1779338 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1779343 - Posted: 15 Apr 2016, 17:34:10 UTC

Now I'm back upstairs, I can see that SoG is still using a heck of a lot of CPU: current display of "CPU efficiency" in BoincView is 0.9788, or 97.88% of a core - pretty good for a CPU application, but compare with ~22% for GPUGrid cuda65, ~11% for SETI cuda50, or <2% for Einstein intel_gpu tasks.

Note that CPU efficiency is (some form of) direct measurement: I think BoincTasks has something similar, though I don't know exactly how either of them work. But they're certainly more realistic than BOINC Manager's echo back of whatever value is written in app_info.xml or estimated by the server.
ID: 1779343 · 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 1779352 - Posted: 15 Apr 2016, 18:02:18 UTC - in response to Message 1779343.  
Last modified: 15 Apr 2016, 18:23:19 UTC

Yes, I use BoincTask as well and they all use 97% of a core each.

It used to be running more than 1 per card reduced that amount but I just finished testin 2 at a time on the cards and there is no difference.

About to try 3 at a time and will see if that reduces any time to complete or % of CPU.

If neither of those happen then I may run r3366 again to see how it compares.
ID: 1779352 · 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 1779371 - Posted: 15 Apr 2016, 19:43:17 UTC - in response to Message 1779352.  

So 3 per Card generates 36-37 minutes, about 1 minute faster than running a single Guppi by itself on the GPU

CPU utilization is down anywhere from 80-92% of a CPU rather than 97%

Only problem is, this require large number of CPU cores to make this happen.
ID: 1779371 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51478
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1779374 - Posted: 15 Apr 2016, 19:46:40 UTC - in response to Message 1779371.  

So 3 per Card generates 36-37 minutes, about 1 minute faster than running a single Guppi by itself on the GPU

CPU utilization is down anywhere from 80-92% of a CPU rather than 97%

Only problem is, this require large number of CPU cores to make this happen.

Not looking very good, is it?
I gotz GPU power, but CPU power is a limitation.
This may be a hard nut to crack for the optimization crew.
"Time is simply the mechanism that keeps everything from happening all at once."

ID: 1779374 · Report as offensive
Previous · 1 . . . 7 · 8 · 9 · 10 · 11 · 12 · 13 . . . 18 · Next

Message boards : Number crunching : OpenCL NV MultiBeam v8 SoG edition for Windows


 
©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.