Ubuntu and Nvidia tasks

Message boards : Number crunching : Ubuntu and Nvidia tasks
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · Next

AuthorMessage
duffy

Send message
Joined: 21 Apr 02
Posts: 4
Credit: 20,448,302
RAC: 0
United States
Message 1364890 - Posted: 5 May 2013, 18:40:40 UTC

I am trying out Ubuntu and have got Seti running on it but can't seem to get any tasks for my GPU. Is it that there are no actual tasks that can be used for this combination? I have been running for a couple of days.

Sun 05 May 2013 02:18:09 PM EDT | | Starting BOINC client version 7.0.27 for i686-pc-linux-gnu
Sun 05 May 2013 02:18:09 PM EDT | | log flags: file_xfer, sched_ops, task
Sun 05 May 2013 02:18:09 PM EDT | | Libraries: libcurl/7.22.0 OpenSSL/1.0.1 zlib/1.2.3.4 libidn/1.23 librtmp/2.3
Sun 05 May 2013 02:18:09 PM EDT | | Data directory: /var/lib/boinc-client
Sun 05 May 2013 02:18:09 PM EDT | | Processor: 2 GenuineIntel Intel(R) Pentium(R) Dual CPU E2200 @ 2.20GHz [Family 6 Model 15 Stepping 13]
Sun 05 May 2013 02:18:09 PM EDT | | Processor: 1.00 MB cache
Sun 05 May 2013 02:18:09 PM EDT | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc arch_perfmon pebs bts aperfmperf pni dtes64 monitor ds_cpl est tm2 ssse3 cx16 xtpr pdcm lahf_lm dtherm
Sun 05 May 2013 02:18:09 PM EDT | | OS: Linux: 3.5.0-28-generic
Sun 05 May 2013 02:18:09 PM EDT | | Memory: 1001.64 MB physical, 1022.00 MB virtual
Sun 05 May 2013 02:18:09 PM EDT | | Disk: 34.90 GB total, 28.04 GB free
Sun 05 May 2013 02:18:09 PM EDT | | Local time is UTC -4 hours
Sun 05 May 2013 02:18:09 PM EDT | | NVIDIA GPU 0: GeForce 210 (driver version unknown, CUDA version 5.0, compute capability 1.2, 512MB, 326MB available, 59 GFLOPS peak)
Sun 05 May 2013 02:18:09 PM EDT | | OpenCL: NVIDIA GPU 0: GeForce 210 (driver version 304.88, device version OpenCL 1.0 CUDA, 512MB, 326MB available)
Sun 05 May 2013 02:18:09 PM EDT | | Config: GUI RPC allowed from:
Sun 05 May 2013 02:18:09 PM EDT | | A new version of BOINC is available. <a href=http://boinc.berkeley.edu/download.php>Download it.</a>
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 6985876; resource share 100
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | General prefs: from SETI@home (last modified 29-Dec-2012 09:58:46)
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | Computer location: home
Sun 05 May 2013 02:18:09 PM EDT | | General prefs: using separate prefs for home
Sun 05 May 2013 02:18:09 PM EDT | | Reading preferences override file
Sun 05 May 2013 02:18:09 PM EDT | | Preferences:
Sun 05 May 2013 02:18:09 PM EDT | | max memory usage when active: 500.82MB
Sun 05 May 2013 02:18:09 PM EDT | | max memory usage when idle: 901.47MB
Sun 05 May 2013 02:18:09 PM EDT | | max disk usage: 10.00GB
Sun 05 May 2013 02:18:09 PM EDT | | suspend work if non-BOINC CPU load exceeds 25 %
Sun 05 May 2013 02:18:09 PM EDT | | (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
Sun 05 May 2013 02:18:09 PM EDT | | Not using a proxy
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | Restarting task 31ja13ab.26341.18633.7.11.200_0 using setiathome_enhanced version 603 in slot 1
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | Restarting task 31ja13ae.28154.17826.3.11.104_0 using setiathome_enhanced version 603 in slot 0
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | Sending scheduler request: To fetch work.
Sun 05 May 2013 02:18:09 PM EDT | SETI@home | Requesting new tasks for NVIDIA
Sun 05 May 2013 02:18:11 PM EDT | SETI@home | Scheduler request completed: got 0 new tasks
Sun 05 May 2013 02:18:11 PM EDT | SETI@home | Project has no tasks available
Sun 05 May 2013 02:27:23 PM EDT | SETI@home | Sending scheduler request: To fetch work.
Sun 05 May 2013 02:27:23 PM EDT | SETI@home | Requesting new tasks for NVIDIA
Sun 05 May 2013 02:27:25 PM EDT | SETI@home | Scheduler request completed: got 0 new tasks
Sun 05 May 2013 02:30:29 PM EDT | SETI@home | Computation for task 31ja13ab.26341.18633.7.11.200_0 finished
Sun 05 May 2013 02:30:29 PM EDT | SETI@home | Starting task 31ja13ab.26341.18633.7.11.170_0 using setiathome_enhanced version 603 in slot 1
Sun 05 May 2013 02:30:31 PM EDT | SETI@home | Started upload of 31ja13ab.26341.18633.7.11.200_0_0
Sun 05 May 2013 02:30:35 PM EDT | SETI@home | Finished upload of 31ja13ab.26341.18633.7.11.200_0_0

ID: 1364890 · 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: 22160
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1364896 - Posted: 5 May 2013, 18:51:42 UTC

There is a shortage of tasks just now, and has been for a week or more :-(
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1364896 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 1364911 - Posted: 5 May 2013, 19:06:24 UTC

Your biggest problem is there is no stock app for Linux CUDA, you will have to go to the anonymous platform and optimized apps to run them.

ID: 1364911 · Report as offensive
Profile Didi

Send message
Joined: 1 Aug 00
Posts: 7
Credit: 45,213,771
RAC: 0
Message 1386859 - Posted: 2 Jul 2013, 19:45:03 UTC

It seems, that I have the same problem... but it is 2 months later.
Is it still the same problem? Is it again the same problem? Or is it another one?
ID: 1386859 · Report as offensive
JohnDK Crowdfunding Project Donor*Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 28 May 00
Posts: 1222
Credit: 451,243,443
RAC: 1,127
Denmark
Message 1386873 - Posted: 2 Jul 2013, 20:37:47 UTC

There haven't been any stock GPU app for Linux in ages, maybe never? There's one now but only for AP tasks, just remember to accept AP6 work in settings.
ID: 1386873 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 1386879 - Posted: 2 Jul 2013, 20:55:26 UTC - in response to Message 1386859.  

It seems, that I have the same problem... but it is 2 months later.
Is it still the same problem? Is it again the same problem? Or is it another one?


Combination of 2 different problems, there is no stock app currently and we just updated to different apps for Multibeam.

But I have updated the app_info for the X41g CUDA app to allow for v7 crunching and included the stock CPU app as well.

http://www.arkayn.us/forum/index.php?action=tpmod;dl=item24

ID: 1386879 · Report as offensive
geyser

Send message
Joined: 7 Oct 04
Posts: 8
Credit: 64,645,821
RAC: 201
United States
Message 1386970 - Posted: 3 Jul 2013, 1:22:40 UTC

After each of my Linux upgrades I always worry that my Nvidia driver does not match my kernel and so my existing GPU tasks fail out. I use this command to monitor the log file:
less -M -i -p gpu\|nvidia /var/lib/boinc/stdoutdae.txt

and then use the follow command, F, to see what is happening. I leave leave it running in a window as a monitor.
I run primegrid and milkeyway because they have native Linux CUDA support.
I was happy with the recent release of the CUDA program for Seti.
Thanks for the expanded Linux support.
ID: 1386970 · Report as offensive
Profile Didi

Send message
Joined: 1 Aug 00
Posts: 7
Credit: 45,213,771
RAC: 0
Message 1387779 - Posted: 4 Jul 2013, 19:44:40 UTC - in response to Message 1386879.  

Thanks.
Now it is loading CUDA-Work. But after one second it's finished with an error.

The messages say:
Don 04 Jul 2013 21:14:19 CEST | | Starting BOINC client version 7.0.65 for x86_64-pc-linux-gnu
Don 04 Jul 2013 21:14:19 CEST | | log flags: file_xfer, sched_ops, task
Don 04 Jul 2013 21:14:19 CEST | | Libraries: libcurl/7.29.0 OpenSSL/1.0.1c zlib/1.2.7 libidn/1.25 librtmp/2.3
Don 04 Jul 2013 21:14:19 CEST | | Data directory: /var/lib/boinc-client
Don 04 Jul 2013 21:14:19 CEST | | Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q8400 @ 2.66GHz [Family 6 Model 23 Stepping 10]
Don 04 Jul 2013 21:14:19 CEST | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts nopl aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm dtherm tpr_shadow vnmi flexpriority
Don 04 Jul 2013 21:14:19 CEST | | OS: Linux: 3.8.0-25-generic
Don 04 Jul 2013 21:14:19 CEST | | Memory: 3.86 GB physical, 149.02 GB virtual
Don 04 Jul 2013 21:14:19 CEST | | Disk: 1.50 TB total, 1.42 TB free
Don 04 Jul 2013 21:14:19 CEST | | Local time is UTC +2 hours
Don 04 Jul 2013 21:14:19 CEST | | CUDA: NVIDIA GPU 0: GeForce GTX 660 (driver version unknown, CUDA version 5.50, compute capability 3.0, 2048MB, 1708MB available, 1982 GFLOPS peak)
Don 04 Jul 2013 21:14:19 CEST | | OpenCL: NVIDIA GPU 0: GeForce GTX 660 (driver version 319.32, device version OpenCL 1.1 CUDA, 2048MB, 1708MB available, 1982 GFLOPS peak)
Don 04 Jul 2013 21:14:19 CEST | SETI@home | Found app_info.xml; using anonymous platform
Don 04 Jul 2013 21:14:19 CEST | | Config: GUI RPCs allowed from:
Don 04 Jul 2013 21:14:19 CEST | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7031125; resource share 100
Don 04 Jul 2013 21:14:19 CEST | SETI@home | General prefs: from SETI@home (last modified 07-Jun-2013 18:50:34)
Don 04 Jul 2013 21:14:19 CEST | SETI@home | Computer location: home
Don 04 Jul 2013 21:14:19 CEST | | General prefs: using separate prefs for home
Don 04 Jul 2013 21:14:19 CEST | | Reading preferences override file
Don 04 Jul 2013 21:14:19 CEST | | Preferences:
Don 04 Jul 2013 21:14:19 CEST | | max memory usage when active: 1976.93MB
Don 04 Jul 2013 21:14:19 CEST | | max memory usage when idle: 3558.47MB
Don 04 Jul 2013 21:14:19 CEST | | max disk usage: 2.00GB
Don 04 Jul 2013 21:14:19 CEST | | (to change preferences, visit a project web site or select Preferences in the Manager)
Don 04 Jul 2013 21:14:19 CEST | | gui_rpc_auth.cfg is empty - no GUI RPC password protection
Don 04 Jul 2013 21:14:19 CEST | | Not using a proxy
Don 04 Jul 2013 21:14:19 CEST | | Suspending GPU computation - user request
Don 04 Jul 2013 21:14:41 CEST | | Resuming GPU computation
Don 04 Jul 2013 21:14:41 CEST | SETI@home | Starting task 27au08ah.8940.7843.10.12.82_0 using setiathome_v7 version 700 (cuda32) in slot 0
Don 04 Jul 2013 21:14:43 CEST | SETI@home | Computation for task 27au08ah.8940.7843.10.12.82_0 finished
Don 04 Jul 2013 21:14:43 CEST | SETI@home | Output file 27au08ah.8940.7843.10.12.82_0_0 for task 27au08ah.8940.7843.10.12.82_0 absent
Don 04 Jul 2013 21:14:43 CEST | SETI@home | Starting task 27au08ah.8940.7843.10.12.123_0 using setiathome_v7 version 700 (cuda32) in slot 0
Don 04 Jul 2013 21:14:44 CEST | SETI@home | Computation for task 27au08ah.8940.7843.10.12.123_0 finished
Don 04 Jul 2013 21:14:44 CEST | SETI@home | Output file 27au08ah.8940.7843.10.12.123_0_0 for task 27au08ah.8940.7843.10.12.123_0 absent
Don 04 Jul 2013 21:14:44 CEST | SETI@home | Starting task 27au08ah.8940.7843.10.12.175_1 using setiathome_v7 version 700 (cuda32) in slot 0
Don 04 Jul 2013 21:14:45 CEST | SETI@home | Computation for task 27au08ah.8940.7843.10.12.175_1 finished
Don 04 Jul 2013 21:14:45 CEST | SETI@home | Output file 27au08ah.8940.7843.10.12.175_1_0 for task 27au08ah.8940.7843.10.12.175_1 absent
Don 04 Jul 2013 21:22:16 CEST | SETI@home | Sending scheduler request: To fetch work.
Don 04 Jul 2013 21:22:16 CEST | SETI@home | Reporting 3 completed tasks
Don 04 Jul 2013 21:22:16 CEST | SETI@home | Requesting new tasks for CPU and NVIDIA
Don 04 Jul 2013 21:22:19 CEST | SETI@home | Scheduler request completed: got 0 new tasks
Don 04 Jul 2013 21:22:19 CEST | SETI@home | No tasks sent
Don 04 Jul 2013 21:22:19 CEST | SETI@home | No tasks are available for SETI@home Enhanced
Don 04 Jul 2013 21:22:19 CEST | SETI@home | No tasks are available for SETI@home v7
Don 04 Jul 2013 21:22:19 CEST | SETI@home | Nachricht vom Server: Ihre app_info.xml Datei enthält keine nutzbare Version von SETI@home v7.
Don 04 Jul 2013 21:22:19 CEST | SETI@home | This computer has finished a daily quota of 1 tasks
Don 04 Jul 2013 21:28:26 CEST | SETI@home | Sending scheduler request: To fetch work.
Don 04 Jul 2013 21:28:26 CEST | SETI@home | Requesting new tasks for CPU
Don 04 Jul 2013 21:28:28 CEST | SETI@home | Scheduler request completed: got 0 new tasks
Don 04 Jul 2013 21:28:28 CEST | SETI@home | No tasks sent
Don 04 Jul 2013 21:28:28 CEST | SETI@home | No tasks are available for SETI@home Enhanced
Don 04 Jul 2013 21:28:28 CEST | SETI@home | No tasks are available for SETI@home v7
Don 04 Jul 2013 21:28:28 CEST | SETI@home | Nachricht vom Server: Ihre app_info.xml Datei enthält keine nutzbare Version von SETI@home v7.
Don 04 Jul 2013 21:28:28 CEST | SETI@home | This computer has finished a daily quota of 1 tasks


Ihre app_info.xml Datei enthält keine nutzbare Version von SETI@home v7.
Why doesn't my app_info.xml contain a usable Version of Seti V7?
I use the "app_info.xml-cuda_and_stock_cpu_v6andv7"...
ID: 1387779 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1387785 - Posted: 4 Jul 2013, 20:17:57 UTC - in response to Message 1387779.  
Last modified: 4 Jul 2013, 20:21:39 UTC

Thanks.
Now it is loading CUDA-Work. But after one second it's finished with an error.

http://setiathome.berkeley.edu/; Computer ID 7031125; resource share 100

Don 04 Jul 2013 21:28:28 CEST | SETI@home | This computer has finished a daily quota of 1 tasks

Not very helpful that you're got your hosts hidden, good job you supplied your host ID.

The Error that Boinc reports is:

<core_client_version>7.0.65</core_client_version>
<![CDATA[
<message>
process exited with code 22 (0x16, -234)
</message>
<stderr_txt>
execv: Permission denied

</stderr_txt>
]]>

Have you set your application permissions?

app_info.xml Datei enthält keine nutzbare Version von SETI@home v7.
Why doesn't my app_info.xml contain a usable Version of Seti V7?
I use the "app_info.xml-cuda_and_stock_cpu_v6andv7"...

That is a Red Herring, you're errored your SETI@home v7 (anonymous platform, nvidia GPU) Wu's out, so your Max Task per Day is only one, the project isn't able to send you any SETI@home v7 (anonymous platform, nvidia GPU) work anyway:

Application details for host 7031125

Claggy
ID: 1387785 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 1387794 - Posted: 4 Jul 2013, 20:41:52 UTC - in response to Message 1387785.  

Also make certain that the execute bit is set as well.

ID: 1387794 · Report as offensive
Profile Didi

Send message
Joined: 1 Aug 00
Posts: 7
Credit: 45,213,771
RAC: 0
Message 1388802 - Posted: 8 Jul 2013, 10:55:23 UTC

Now it works. Thank you, for your help!

But it works very slowly and there is no work form the server for the CPU. I Think about runing Seti on Windows, until there is a better solution for Linux.
ID: 1388802 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1388803 - Posted: 8 Jul 2013, 11:06:17 UTC - in response to Message 1388802.  

and there is no work form the server for the CPU.

Did you put the Stock Linux CPU app in your app_info too?

Claggy
ID: 1388803 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1388837 - Posted: 8 Jul 2013, 13:09:20 UTC - in response to Message 1388802.  
Last modified: 8 Jul 2013, 13:42:58 UTC

Now it works. ... But it works very slowly ...

'very slowly' means nothing, give real time to complete the tasks and what 'angle range' are they.
CUDA have to be 'much faster' (e.g. x10 speed) compared to CPU

Edit:
You have one completed 'SETI@home v7 Anonymous platform (NVIDIA GPU)':
http://setiathome.berkeley.edu/result.php?resultid=3064315856

Really have too big Run time 84,087.52 (23 h)

At the start of task: 'clockRate = 1032500' (1032 MHz) but it may change during task run -
check if the GPU don't downclock or go in some power-save mode

I don't use Linux but found this (first result on Google for 'Ubuntu Nvidia power-save mode')
Nvidia Powermizer defaults to adaptive mode:
http://ubuntuforums.org/showthread.php?t=1478192

"Nvidia appears to have kindly designed a power saving mode (Powermizer) for our video cards, by clocking down the GPU"


Same GPU and same app (another user):
http://setiathome.berkeley.edu/results.php?hostid=5643864&offset=0&show_names=0&state=4&appid=11

'Run time 2,092.58', 'clockRate = 1071500':
http://setiathome.berkeley.edu/result.php?resultid=3066908080


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1388837 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1388841 - Posted: 8 Jul 2013, 13:23:41 UTC - in response to Message 1388837.  

Now it works. ... But it works very slowly ...

'very slowly' means nothing, give real time to complete the tasks and what 'angle range' are they.
CUDA have to be 'much faster' (e.g. x10 speed) compared to CPU


How about 84,087 seconds (almost 24 hours) for a Mid Range AR task on a GTX660:

All tasks for computer 7031125

Claggy
ID: 1388841 · Report as offensive
S@NL - John van Gorsel
Volunteer tester
Avatar

Send message
Joined: 5 Jul 99
Posts: 193
Credit: 139,673,078
RAC: 0
Netherlands
Message 1388847 - Posted: 8 Jul 2013, 13:43:21 UTC - in response to Message 1388841.  

Now it works. ... But it works very slowly ...

'very slowly' means nothing, give real time to complete the tasks and what 'angle range' are they.
CUDA have to be 'much faster' (e.g. x10 speed) compared to CPU


How about 84,087 seconds (almost 24 hours) for a Mid Range AR task on a GTX660:

All tasks for computer 7031125

Claggy


I'm running the same GPU application on a GTX560Ti on this host and it takes around 1200 seconds for a mid range AR task (running 2 tasks in parallel).

This host is running an older Linux version, Boinc 7.0.25, and nVidia driver 310.40. Other than that, I don't see significant differences between the stderr outputs.



Seti@Netherlands website
ID: 1388847 · Report as offensive
Profile Didi

Send message
Joined: 1 Aug 00
Posts: 7
Credit: 45,213,771
RAC: 0
Message 1389368 - Posted: 10 Jul 2013, 5:33:05 UTC - in response to Message 1388802.  

The Same Computer under Win XP has Host ID: 4354565
The GPU does the same Work in 18 minuts that needs 24 hours under Linux.
ID: 1389368 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1389420 - Posted: 10 Jul 2013, 9:28:52 UTC - in response to Message 1389368.  

The Same Computer under Win XP has Host ID: 4354565
The GPU does the same Work in 18 minutes that needs 24 hours under Linux.

What stops you to fix the Ubuntu Nvidia Powermizer settings?
(it's not true that 'it' "needs 24 hours under Linux", see other people Linux computers)


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1389420 · Report as offensive
.clair.

Send message
Joined: 4 Nov 04
Posts: 1300
Credit: 55,390,408
RAC: 69
United Kingdom
Message 1389891 - Posted: 12 Jul 2013, 0:19:29 UTC - in response to Message 1389420.  

The Same Computer under Win XP has Host ID: 4354565
The GPU does the same Work in 18 minutes that needs 24 hours under Linux.

What stops you to fix the Ubuntu Nvidia Powermizer settings?
(it's not true that 'it' "needs 24 hours under Linux", see other people Linux computers)

Using Ubu 10.4 lts and nv280 one wu at a time.
Of the tasks that my 8800GTX has crunched and validated the longest running one took 2257 seconds - 37 min
Its an old slow card compared to new kit.
ID: 1389891 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1389917 - Posted: 12 Jul 2013, 1:33:04 UTC - in response to Message 1389891.  


("nv280" is probably the version of Nvidia driver?)

Did you need to make some adjustments/edits to Nvidia driver Powermizer settings?
(I don't know if 'Powermizer' is the current name that Nvidia uses for this)

What mode/profile for GPU power-saving was set by default?


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1389917 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1390223 - Posted: 12 Jul 2013, 19:33:27 UTC
Last modified: 12 Jul 2013, 19:37:21 UTC

From my experience with Ubuntu using recent nVidia GPU drivers, even Multibeam tasks require a free CPU core to work correctly. The MB times were twice as fast with 1 free core as opposed to just setting the 'use at most ___ % CPU time' to around 95% hoping that would be enough. It wasn't enough, it needed 1 complete core free. One of these days I'm going to try to install nVidia Driver 266.56 in Ubuntu and see if that makes the same difference it does in Windows. As for now, with 1 free core, my Ubuntu times are about the same as they are in Windows XP.
ID: 1390223 · Report as offensive
1 · 2 · 3 · Next

Message boards : Number crunching : Ubuntu and Nvidia 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.