Linux CUDA 'Special' App finally available, featuring Low CPU use

Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 60 · 61 · 62 · 63 · 64 · 65 · 66 . . . 71 · Next

AuthorMessage
Profile petri33Project Donor
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1465
Credit: 269,098,910
RAC: 295,349
Finland
Message 1895112 - Posted: 13 Oct 2017, 20:59:14 UTC - in response to Message 1895069.  
Last modified: 13 Oct 2017, 21:01:23 UTC

This is a Typical Invalid Overflow, https://setiathome.berkeley.edu/workunit.php?wuid=2708379644
The way I remember it, the CUDA App looks for Triplets First. If a Task starts with many Triplets the Overflow result will be 30 Triplets.
If the App looks for Something Else First, such as the SoG App, then the results will most likely be 30 of whatever it is looking for, i.e. Not Triplets.
It actually may not be that simple. During an email exchange I had with Petri about a month ago regarding this problem, he said "I looked at my code and the pulses are checked before triplets. So it is not so an easy fix I thought. I will have to debug why my code misses many pulses on noisy packets and then some on 'normal' data."

To complicate it further, it seems to be a problem that already exists in the older Cuda apps, as I noted previously, so it may be in some code that Petri's app actually inherited from the stock Cuda code. It just never surfaced until the 4-bit WUs started to flow.


Thank you Jeff,

There is going to be a major overhaul of the code if the 'old' cuda code base is going ever to be compatible with SoG or any other version regarding to the pulses on overflow packets. The noise would induce an unnecessary lot of work to code in a parallel environment and induce a slowdown for what? An eye candy! !! ! !!!! ! !!!!! !!!!!!!!! !! !!!!!! ...

An overflow is an overflow and any two similar apps can confirm that. If the apps are not the same then it will be asked from an another one if that it is a bad packet indeed. If nonagreement then ask anoher. etc. A 30/30 or a near miss is a noise bomb. BOOM!

No Signature Here.

EDIT: I'm a week off here starting now!
ID: 1895112 · Report as offensive     Reply Quote
Profile Jeff Buck
Volunteer tester

Send message
Joined: 11 Feb 00
Posts: 1272
Credit: 133,699,582
RAC: 241,114
United States
Message 1895122 - Posted: 13 Oct 2017, 21:24:49 UTC - in response to Message 1895112.  

My comment wasn't intended to suggest that the problem didn't need to be fixed. I was just noting that I've seen it appear with the stock Cuda apps, in particular the Cuda50 running on my daily driver. And I was passing along your analysis that it wasn't just a processing order issue.

Ideally, Jason would probably be the one to try to track it down in the current Cuda codebase, but he has been absent for awhile, so if it can be fixed in the Special App, I would expect that it could be ported back to the more widely used Cuda apps.

As I think I've expressed multiple times previously, just because a WU overflows doesn't mean that it's worthless. That 30 signal cutoff was based on storage considerations, not the value of the scientific data. The apps need to report consistently and let the scientists sort through the results and make any "noise bomb" determination. Anyway, what appears to be a noise bomb to one person might actually turn out to be an alien ABBA concert to another. ;^)
ID: 1895122 · Report as offensive     Reply Quote
Profile petri33Project Donor
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1465
Credit: 269,098,910
RAC: 295,349
Finland
Message 1895131 - Posted: 13 Oct 2017, 21:46:08 UTC - in response to Message 1895122.  
Last modified: 13 Oct 2017, 21:46:36 UTC

My comment wasn't intended to suggest that the problem didn't need to be fixed. I was just noting that I've seen it appear with the stock Cuda apps, in particular the Cuda50 running on my daily driver. And I was passing along your analysis that it wasn't just a processing order issue.

Ideally, Jason would probably be the one to try to track it down in the current Cuda codebase, but he has been absent for awhile, so if it can be fixed in the Special App, I would expect that it could be ported back to the more widely used Cuda apps.

As I think I've expressed multiple times previously, just because a WU overflows doesn't mean that it's worthless. That 30 signal cutoff was based on storage considerations, not the value of the scientific data. The apps need to report consistently and let the scientists sort through the results and make any "noise bomb" determination. Anyway, what appears to be a noise bomb to one person might actually turn out to be an alien ABBA concert to another. ;^)


ABBA! Yes. One night in a "pulse nightmare" (an old one "jungle") no.

Choose your game. There are rules. Some hands in a Poker game give you points, some Ace in a sleeve does not. Change the rules - lower the limit to regard a 'signal' that is below the noise lever (i.e. below one scorewise) to be accepted. I could report all of them! To find the best that is not reported is time consuming in a parallel world. Be prepared for the quantum era.

Mamma Mia! No, no. The Winner does not take it all - it is just on the wings of the next hurricane named after a man - Fernando. - ABBA quote off.

Still no name until I fall asleep.
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1895131 · Report as offensive     Reply Quote
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 5805
Credit: 76,005,510
RAC: 50,869
Russia
Message 1895145 - Posted: 13 Oct 2017, 22:31:09 UTC - in response to Message 1895131.  

To find the best that is not reported is time consuming in a parallel world.

Yes. indeed. But still it's part of algorithm.
Until algorithm will be changed best should be found correctly (some sort of reduction from best per CU to single best could be used to reduce slowdown from serialization.
Regarding overflows - yep, early versions of SoG had same issue too. The more distributed task computation is the bigger amount of signals one should store to properly reorder on reporting. At some point it will too costly indeed. But if you still doing one icfft per kernel call amount of signals to keep should be not too huge.
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1895145 · Report as offensive     Reply Quote
Profile petri33Project Donor
Volunteer tester

Send message
Joined: 6 Jun 02
Posts: 1465
Credit: 269,098,910
RAC: 295,349
Finland
Message 1895161 - Posted: 13 Oct 2017, 23:26:25 UTC - in response to Message 1895145.  

To find the best that is not reported is time consuming in a parallel world.

Yes. indeed. But still it's part of algorithm.
Until algorithm will be changed best should be found correctly (some sort of reduction from best per CU to single best could be used to reduce slowdown from serialization.
Regarding overflows - yep, early versions of SoG had same issue too. The more distributed task computation is the bigger amount of signals one should store to properly reorder on reporting. At some point it will too costly indeed. But if you still doing one icfft per kernel call amount of signals to keep should be not too huge.


+1, from the far end of the visible world. I'll be back.
To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones
ID: 1895161 · Report as offensive     Reply Quote
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 3789
Credit: 186,266,825
RAC: 237,427
United States
Message 1895356 - Posted: 14 Oct 2017, 16:24:02 UTC - in response to Message 1895161.  
Last modified: 14 Oct 2017, 16:26:29 UTC

I'll be back.
I'm starting to accumulate quite a few Invalid Overflows on the 750Ti machine. All three of my machines are running the same version of zi3xs2 compiled in Ubuntu 12.04.5, don't know why that machine is getting more Invalids. This last one included a machine running the old CUDA 4.2 App, and it is a bit different than zi3xs2;

24fe07ad.6233.23794.15.42.0
x41p_zi3xs2, Cuda 9.00 special
Find triplets Cuda kernel encountered too many triplets, or bins above threshold, reprocessing this PoT on CPU... err = 1
Triplet: peak=12.57254, time=62.22, period=0.4198, d_freq=1420001220.7, chirp=0, fft_len=8
Pulse: peak=10.81523, time=6.946, period=4.474, d_freq=1419995117.19, score=1.029, chirp=0, fft_len=16
Pulse: peak=9.594464, time=13.89, period=3.077, d_freq=1419995117.19, score=1.169, chirp=0, fft_len=16
Spike: peak=24.94294, time=0.03275, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=28.80101, time=0.09829, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.6429, time=0.1114, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=28.34846, time=0.1245, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.6764, time=0.1769, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=27.97159, time=0.2556, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=25.79398, time=0.3211, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=32.19441, time=0.3342, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.18413, time=0.4129, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=27.33308, time=0.5046, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.42743, time=0.5964, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=30.34237, time=0.6619, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.09592, time=0.7012, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=27.51006, time=0.7668, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=26.7412, time=0.793, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=26.68621, time=0.8061, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=25.5911, time=0.8716, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.99944, time=0.9372, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=27.33264, time=0.9503, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.72494, time=1.003, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=27.52413, time=1.173, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=25.03278, time=1.186, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=24.43941, time=1.199, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=26.34531, time=1.239, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=25.7531, time=1.265, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=25.50618, time=1.409, d_freq=1420000000, chirp=0, fft_len=128
Spike: peak=25.16733, time=1.488, d_freq=1420000000, chirp=0, fft_len=128
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected equals the storage space allocated.

Best spike: peak=32.19441, time=0.3342, d_freq=1420000000, chirp=0, fft_len=128
Best autocorr: peak=0, time=-2.12e+11, delay=0, d_freq=0, chirp=0, fft_len=0
Best gaussian: peak=0, mean=0, ChiSq=0, time=-2.12e+11, d_freq=0,
score=-12, null_hyp=0, chirp=0, fft_len=0
Best pulse: peak=9.594464, time=13.89, period=3.077, d_freq=1419995117.19, score=1.169, chirp=0, fft_len=16
Best triplet: peak=12.57254, time=62.22, period=0.4198, d_freq=1420001220.7, chirp=0, fft_len=8
Spike count: 27
Autocorr count: 0
Pulse count: 2
Triplet count: 1
Gaussian count: 0


x41p_zi3t1d, Cuda 8.00 special
Find triplets Cuda kernel encountered too many triplets, or bins above threshold, reprocessing this PoT on CPU... err = 1
Triplet: peak=12.57254, time=62.22, period=0.4198, d_freq=1420001220.7, chirp=0, fft_len=8
Pulse: peak=10.81523, time=6.946, period=4.474, d_freq=1419995117.19, score=1.029, chirp=0, fft_len=16
Pulse: peak=9.594468, time=13.89, period=3.077, d_freq=1419995117.19, score=1.169, chirp=0, fft_len=16
Pulse: peak=7.708465, time=13.9, period=2.792, d_freq=1419995803.83, score=1.012, chirp=0, fft_len=128
Pulse: peak=8.676339, time=34.74, period=2.805, d_freq=1419995803.83, score=1.139, chirp=0, fft_len=128
Pulse: peak=8.324149, time=41.69, period=2.805, d_freq=1419995803.83, score=1.093, chirp=0, fft_len=128
Pulse: peak=8.365986, time=83.37, period=3.076, d_freq=1419995803.83, score=1.095, chirp=0, fft_len=128
Pulse: peak=7.730057, time=34.75, period=3.067, d_freq=1420003547.67, score=1.038, chirp=0, fft_len=256
Pulse: peak=6.753286, time=13.91, period=2.661, d_freq=1420003585.82, score=1.087, chirp=0, fft_len=256
Pulse: peak=8.121024, time=34.75, period=2.787, d_freq=1420003585.82, score=1.094, chirp=0, fft_len=256
Pulse: peak=8.506479, time=6.96, period=3.102, d_freq=1420004196.17, score=1.141, chirp=0, fft_len=256
Pulse: peak=7.733213, time=13.91, period=2.628, d_freq=1420004196.17, score=1.245, chirp=0, fft_len=256
Pulse: peak=7.576841, time=48.64, period=3.46, d_freq=1420004196.17, score=1.012, chirp=0, fft_len=256
Pulse: peak=6.45851, time=41.69, period=2.438, d_freq=1419995536.8, score=1.042, chirp=0, fft_len=256
Pulse: peak=7.476565, time=34.75, period=3.198, d_freq=1419995765.69, score=1.002, chirp=0, fft_len=256
Pulse: peak=9.763989, time=41.69, period=3.591, d_freq=1419995765.69, score=1.033, chirp=0, fft_len=256
Pulse: peak=8.356166, time=6.96, period=3.102, d_freq=1419996414.18, score=1.121, chirp=0, fft_len=256
Spike: peak=25.77461, time=1.887, d_freq=1420000848.77, chirp=0, fft_len=4k
Spike: peak=25.85038, time=4.404, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=28.45691, time=7.34, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=25.20111, time=8.179, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=26.39505, time=8.598, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=26.16997, time=9.018, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=25.20473, time=9.437, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=30.48201, time=10.28, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=30.76434, time=10.7, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=31.26083, time=11.95, d_freq=1420000848.77, chirp=0, fft_len=4k
Spike: peak=29.29335, time=15.73, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=28.20589, time=17.41, d_freq=1419999151.23, chirp=0, fft_len=4k
Spike: peak=39.77211, time=30.41, d_freq=1419999153.61, chirp=0, fft_len=4k
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected equals the storage space allocated.

Best spike: peak=39.77211, time=30.41, d_freq=1419999153.61, chirp=0, fft_len=4k
Best autocorr: peak=0, time=-2.12e+11, delay=0, d_freq=0, chirp=0, fft_len=0
Best gaussian: peak=0, mean=0, ChiSq=0, time=-2.12e+11, d_freq=0,
score=-12, null_hyp=0, chirp=0, fft_len=0
Best pulse: peak=7.733213, time=13.91, period=2.628, d_freq=1420004196.17, score=1.245, chirp=0, fft_len=256
Best triplet: peak=12.57254, time=62.22, period=0.4198, d_freq=1420001220.7, chirp=0, fft_len=8
Spike count: 13
Autocorr count: 0
Pulse count: 16
Triplet count: 1
Gaussian count: 0


x41zi (baseline v8), Cuda 4.20
Find triplets Cuda kernel encountered too many triplets, or bins above threshold, reprocessing this PoT on CPU...
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected equals the storage space allocated.
Spike count: 13
Autocorr count: 0
Pulse count: 16
Triplet count: 1
Gaussian count: 0
ID: 1895356 · Report as offensive     Reply Quote
Stephen "Heretic"Project Donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 2630
Credit: 48,102,961
RAC: 131,121
Australia
Message 1895428 - Posted: 15 Oct 2017, 1:18:52 UTC - in response to Message 1887088.  


I have a new version which can :
1) move tasks from gpu to cpu (-r flag)
2) move only suspended tasks (-s flag).

If you want to have a try....


. . 8^} Where do ya get it ??

Stephen

:)


Sorry, you can get it at :
http://www.w3perl.com/seti/cpu2gpu.pl

beta software ! Let me know how to improve or send me bug report ! ;)


. . Hi Laurent,

. . I have not given any feedback on your revision of your app. It is working very very well and I am having great success with it. It is very fast in terminal window.

Stephen

:)
ID: 1895428 · Report as offensive     Reply Quote
Bruce
Volunteer tester

Send message
Joined: 15 Mar 02
Posts: 94
Credit: 91,592,353
RAC: 53,883
United States
Message 1895461 - Posted: 15 Oct 2017, 5:17:05 UTC - in response to Message 1895056.  

So far the version of zi3xs2 compiled in Ubuntu 12.04.5 seems to be working normally. It still has the same occasional Bad Best Pulse it's had Forever, but, the biggest problem is all the Inconclusive Instant Overflows which account for over half the inconclusive count. Ever so often one of the Instant Overflows is given an Invalid. If those Overflow Inconclusives could be reduced it would look Much better. I haven't been able to compile it in anything other than CUDA 9, so, if the CUDA 9 version doesn't work on the Kepler cc 3.5 GPUs there isn't much hope for them.

Hi TBar,
I would be willing to try the new s2 version.
How would I need to change things? Will I need the Cuda 9 libraries?
Newer driver not a problem to install.
Any help appreciated.
If the new cuda apps are not going to work with older hardware, we need to know.
Thanks.
Bruce
ID: 1895461 · Report as offensive     Reply Quote
Profile Keith Myers
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 2428
Credit: 183,918,647
RAC: 355,199
United States
Message 1895462 - Posted: 15 Oct 2017, 5:26:00 UTC - in response to Message 1895461.  
Last modified: 15 Oct 2017, 5:27:08 UTC

Back in Petri's original post he stated. This executable is only for those who can test offline on L I N U X and GTX10x0 !!! sm_61 i.e. GTX10x0 only !!!.. You don't need to download the CUDA 9.0 libraries as the executable has them statically linked internally. You would also need the latest drivers with CUDA 9.0 support.
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1895462 · Report as offensive     Reply Quote
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 3789
Credit: 186,266,825
RAC: 237,427
United States
Message 1895475 - Posted: 15 Oct 2017, 7:46:42 UTC - in response to Message 1895461.  

Hi TBar,
I would be willing to try the new s2 version.
How would I need to change things? Will I need the Cuda 9 libraries?
Newer driver not a problem to install.
Any help appreciated.
If the new cuda apps are not going to work with older hardware, we need to know.
Thanks.
Check your PMs and let me know how that works.
ID: 1895475 · Report as offensive     Reply Quote
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 3789
Credit: 186,266,825
RAC: 237,427
United States
Message 1895487 - Posted: 15 Oct 2017, 12:02:54 UTC

If anyone else has a Kepler cc 3.5 GPU post a link to the machine and I'll give you a link to the zi3xs2-35 App. I really don't see any difference with this App verses the ones compiled with the cc 5.0 & 5.2 flags on my machines. The first time you run it on a different GPU it will take about 20 seconds to compile and cache the correct GPU code, after that first time the run times are about the same as an App compiled with the different sm flags. It should work on Kepler cc 3.5 and higher GPUs in Ubuntu 12.04.5 and higher with driver 384.xx. The objective is to see if CUDA 9 will work on the cc 3.5 GPUs.
ID: 1895487 · Report as offensive     Reply Quote
Stephen "Heretic"Project Donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 2630
Credit: 48,102,961
RAC: 131,121
Australia
Message 1895488 - Posted: 15 Oct 2017, 12:35:03 UTC - in response to Message 1895487.  

If anyone else has a Kepler cc 3.5 GPU post a link to the machine and I'll give you a link to the zi3xs2-35 App. I really don't see any difference with this App verses the ones compiled with the cc 5.0 & 5.2 flags on my machines. The first time you run it on a different GPU it will take about 20 seconds to compile and cache the correct GPU code, after that first time the run times are about the same as an App compiled with the different sm flags. It should work on Kepler cc 3.5 and higher GPUs in Ubuntu 12.04.5 and higher with driver 384.xx. The objective is to see if CUDA 9 will work on the cc 3.5 GPUs.


. . Hi TBar

. . OK, here's the thing ... I do have a Kepler CC3.5 card (GT730) but the only spare slot I can put it into is the i5-6600 with the 2 x 970s.

. . So I don't think that would be a viable setup. The 970s were running fine with a GTX1050 in that slot until the weather got too warm but I don't think that a GT730 would play well with the 970s. If you feel otherwise let me know and I will give it a try.

. . The only other alternative is to remove the 1050ti from the Core2 Duo and put it in there. That might be a viable test but would the S2 version be OK when I return to the 1050ti??

Stephen

??
ID: 1895488 · Report as offensive     Reply Quote
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 3789
Credit: 186,266,825
RAC: 237,427
United States
Message 1895525 - Posted: 15 Oct 2017, 17:38:17 UTC - in response to Message 1895488.  

...but would the S2 version be OK when I return to the 1050ti??
If it wasn't, what would stop you from going back to what you are using now???
Have you ever run the 730 with the Existing CUDA 8 & 6.5 Apps? If you decide to try the newer App, it would be nice to be able to compare it to the Existing Apps.
ID: 1895525 · Report as offensive     Reply Quote
Profile Keith Myers
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 2428
Credit: 183,918,647
RAC: 355,199
United States
Message 1895538 - Posted: 15 Oct 2017, 19:00:58 UTC

Are we still in pre-production testing for the CUDA 9.0 apps? Or are they usable in production now?
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1895538 · Report as offensive     Reply Quote
W3PerlProject Donor
Volunteer tester

Send message
Joined: 29 Apr 99
Posts: 153
Credit: 957,832,418
RAC: 2,247,595
France
Message 1895543 - Posted: 15 Oct 2017, 19:20:32 UTC - in response to Message 1895428.  



. . Hi Laurent,

. . I have not given any feedback on your revision of your app. It is working very very well and I am having great success with it. It is very fast in terminal window.

Stephen

:)


Thanks for the report :)
Glad to know it could be useful to you.
I have updated the script once again to fix some issues, you can retrieve it here
ID: 1895543 · Report as offensive     Reply Quote
Stephen "Heretic"Project Donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 2630
Credit: 48,102,961
RAC: 131,121
Australia
Message 1895569 - Posted: 15 Oct 2017, 21:30:31 UTC - in response to Message 1895525.  

...but would the S2 version be OK when I return to the 1050ti??
If it wasn't, what would stop you from going back to what you are using now???
Have you ever run the 730 with the Existing CUDA 8 & 6.5 Apps? If you decide to try the newer App, it would be nice to be able to compare it to the Existing Apps.


. . Sadly I replaced the 730 with the 1050ti before the general release of the CUDA6.5/8.0. I can give you comparisons with SoG but not with earlier versions of the special sauce.

. . I probably still have the downloads of an earlier version on that machine if you need a comparison base.

. . And you didn't give a response to the question of which setup you think I should try. Though it is my impression it is the second option that you think is most viable.

Stephen

??
ID: 1895569 · Report as offensive     Reply Quote
Stephen "Heretic"Project Donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 2630
Credit: 48,102,961
RAC: 131,121
Australia
Message 1895574 - Posted: 15 Oct 2017, 22:08:32 UTC - in response to Message 1895525.  

...but would the S2 version be OK when I return to the 1050ti??
If it wasn't, what would stop you from going back to what you are using now???
Have you ever run the 730 with the Existing CUDA 8 & 6.5 Apps? If you decide to try the newer App, it would be nice to be able to compare it to the Existing Apps.


. . I have looked on this machine and I still have downloads for 3k+ and 3t2b.

Stephen

??
ID: 1895574 · Report as offensive     Reply Quote
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 3789
Credit: 186,266,825
RAC: 237,427
United States
Message 1895579 - Posted: 15 Oct 2017, 22:42:26 UTC - in response to Message 1895574.  
Last modified: 15 Oct 2017, 23:08:24 UTC

I would recommend the Machine with the single GPU. It's usually best to test with just one GPU, and if you wish you could place the 1050Ti in the Multi-GPU machine. Use the current CUDA Apps, you can start with what's already installed. It should be a simple matter of just suspending all the GPU tasks in BOINC and then shutdown the machine and swap the cards. Once running just resume One GPU task and see how it works with the 730. If it works, run a few more and then Swap the CUDA Apps. I think someone said the CUDA 6.5 App is demanding the CUDA 6.5 Libraries instead of the 6.0 Libraries in the Links. I'll see if I can Upload the 6.5 Libraries to C.A. so you don't have to download the 6.5 ToolKit. To change the Apps just paste the App & Libraries into the setiathome.berkeley.edu folder and change the app_info.xml to name the New App & Libraries. The Library names don't matter to the CUDA 9 App, it will just use the Built-in Libraries anyway.

The Link is Up for the 6.5 Libraries at C.A. http://www.arkayn.us/forum/index.php?topic=197.msg4499#msg4499
ID: 1895579 · Report as offensive     Reply Quote
Stephen "Heretic"Project Donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 2630
Credit: 48,102,961
RAC: 131,121
Australia
Message 1895588 - Posted: 15 Oct 2017, 23:48:54 UTC - in response to Message 1895579.  

I would recommend the Machine with the single GPU. It's usually best to test with just one GPU, and if you wish you could place the 1050Ti in the Multi-GPU machine. Use the current CUDA Apps, you can start with what's already installed. It should be a simple matter of just suspending all the GPU tasks in BOINC and then shutdown the machine and swap the cards. Once running just resume One GPU task and see how it works with the 730. If it works, run a few more and then Swap the CUDA Apps. I think someone said the CUDA 6.5 App is demanding the CUDA 6.5 Libraries instead of the 6.0 Libraries in the Links. I'll see if I can Upload the 6.5 Libraries to C.A. so you don't have to download the 6.5 ToolKit. To change the Apps just paste the App & Libraries into the setiathome.berkeley.edu folder and change the app_info.xml to name the New App & Libraries. The Library names don't matter to the CUDA 9 App, it will just use the Built-in Libraries anyway.

The Link is Up for the 6.5 Libraries at C.A. http://www.arkayn.us/forum/index.php?topic=197.msg4499#msg4499


. . OK, that would save time, but I was in the process of running down the cached tasks. I am running 3v at the moment so is there any issue with that and Kepler cc3.5 cards? I seem to remember people having issues with GTX780 units.

. . And that is a good idea of putting the 1050ti into the i5 rig. It is low profile and should not restrict airflow as much as the 1050 I had in there before.

Stephen

??
ID: 1895588 · Report as offensive     Reply Quote
Profile Keith Myers
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 2428
Credit: 183,918,647
RAC: 355,199
United States
Message 1895589 - Posted: 15 Oct 2017, 23:55:00 UTC - in response to Message 1895538.  

Are we still in pre-production testing for the CUDA 9.0 apps? Or are they usable in production now?

[bump]
Seti@Home classic workunits:20,676 CPU time:74,226 hours
ID: 1895589 · Report as offensive     Reply Quote
Previous · 1 . . . 60 · 61 · 62 · 63 · 64 · 65 · 66 . . . 71 · Next

Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use


 
©2017 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.