Ubuntu Server 12.04.4 - Use GPU to chrunch data


log in

Advanced search

Questions and Answers : GPU applications : Ubuntu Server 12.04.4 - Use GPU to chrunch data

Author Message
DanHansen@Denmark
Volunteer tester
Avatar
Send message
Joined: 14 Nov 12
Posts: 162
Credit: 2,056,165
RAC: 12,054
Denmark
Message 1478001 - Posted: 16 Feb 2014, 15:00:10 UTC

Hi crunchers,

I've worked through the hours to find a solution on this problem, and apparently I'm not the only one:

To use the GPU in a Nvidia GeForce GT610 to crunch data with the boinc-client. The boinc-client because we are in a non-graphical environment with RACK mounted Servers.

Several of you guys in here, has really done a lot to fix this problem, but we are not there yet. I recently found a very good site, where the solution might be. Please read this URL and if possible, translate it to "Ubuntu Server Language", maybe a kind of ToDo!?

Project: Ubuntu Server 12.04.4 64bit CPU & GPU Boinc Data Cruncher
CPU: Intel i5-3470 processor
GPU: Nvidia GeForce GT 610 (will be replaced by GT640)

Please keep in mind, that what we are looking for, is a solution where we use Ubuntu Server 12.04.4 a non-graphical environment. RACK mounted servers, all in the usual shell mode, therefore no boinc-manager, only the boinc-client!
Here's the link for the best possible solution for this matter:
http://ubuntuforums.org/showthread.php?t=1131430
http://www.hardwarecanucks.com/forum/hardwarecanucks-f-h-team/58582-experiment-whats-minimal-install-get-nvidia-boinc-crunching.html
http://www.gpugrid.net/forum_thread.php?id=1401
http://www.gpugrid.net/forum_thread.php?id=3533
http://www.gpugrid.net/forum_thread.php?id=3083

I'll call upon all you BOINC Guru's out there, to join in and get this done ;)

I will keep trying, keep wiping and reinstalling OS, drivers etc. in the name of serious data crunching, looking for little green men and large stones in the air ;)

.
____________
Keep On Crunching!

Project RACK Linux Boinc Servers
Ubuntu Server 12.04.4 64bit
Intel i5/8Gb/P8H77-M
Asus GeForceGT640 2Gb
Nvidia v.nvidia-linux-x86_64-331.38
LM-Sensors v.3.3.1 with libsensors version 3.3.1
BOINC v.7.2.33 x86_64-pc-linux-gn

DanHansen@Denmark
Volunteer tester
Avatar
Send message
Joined: 14 Nov 12
Posts: 162
Credit: 2,056,165
RAC: 12,054
Denmark
Message 1478298 - Posted: 17 Feb 2014, 11:35:47 UTC
Last modified: 17 Feb 2014, 11:41:12 UTC

Hi,

Boinc: 7.2.33
Ubuntu Server 12.04.4 64 bit
Nvidia: v.nvidia-linux-x86_64-331.38

Here's the proof. Clean installation of Ubuntu Server 12.04.2, then GCC, blacklisting of the nouveau, installation of make, then downloading and installation of Nvidia driver.

GPU recognized and can be seen in "http://setiathome.berkeley.edu/hosts_user.php".

But this is the results of the first 10 minutes. Same result. It looks like it work, but it doesn't.:



Clean installation of Ubuntu Server 12.04.4 - blacklisted nouveau and others - install gcc - install make - install driver - install boinc-client. See todo in the bottom:


Right after new installation:


# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 34 C

# boinccmd --get_tasks

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 0.585386
current CPU time: 0.636268 <-- IS IT RUNNING?
fraction done: 0.002143
swap size: 17325019136.000000
working set size: 24019928.294922
estimated CPU time remaining: 46310.397652 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!


About 30 sec. later:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 38 C

# boinccmd --get_tasks

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 0.698044
current CPU time: 178.823500 <-- STILL RUNNING OR SO IT SEEMS AT LEAST!
fraction done: 0.004049
swap size: 17325019136.000000
working set size: 24027135.983633
estimated CPU time remaining: 46088.568553 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!



After a couple of minutes:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 35 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 0.698044
current CPU time: 419.934200 <-- STILL RUNNING OR SO IT SEEMS AT LEAST!
fraction done: 0.004049
swap size: 17325019136.000000
working set size: 24027136.000000
estimated CPU time remaining: 45848.038281 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!


After about 3 min.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 34 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 0.698044
current CPU time: 714.048400 <-- STILL RUNNING OR SO IT SEEMS AT LEAST!
fraction done: 0.004049
swap size: 17325019136.000000
working set size: 24027136.000000
estimated CPU time remaining: 45555.583161 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!



****************************************************************************
AFTER REBOOT - IT LOOKS LIKE ITS STARTING OVER. LOOK AT THE CURRENT CPU-TIME

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 35 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 0.698044
current CPU time: 1.010883 <-- IS IT RUNNING?
fraction done: 0.004049
swap size: 17324924928.000000
working set size: 23929983.999999
estimated CPU time remaining: 46193.601236 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!


After 30 sek.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 39 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.159562
current CPU time: 1.265546 <-- IS IT RUNNING?
fraction done: 0.007859
swap size: 17325019136.000000
working set size: 23998459.820312
estimated CPU time remaining: 46051.810281 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!


After 1 min.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 40 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.296961
current CPU time: 1.395721 <-- IS IT RUNNING?
fraction done: 0.009764
swap size: 17325019136.000000
working set size: 23998463.967346
estimated CPU time remaining: 45978.306833 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!


After 2 min.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 41 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.431833
current CPU time: 1.502592 <-- IS IT RUNNING?
fraction done: 0.011669
swap size: 17325019136.000000
working set size: 23998463.999490
estimated CPU time remaining: 45916.704743 <-- LOOKS LIKE IT'S RUNNING OR SO IT SEEMS AT LEAST!


After 5min.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 42 C

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.705611
current CPU time: 1.825205
fraction done: 0.015480
swap size: 17325019136.000000
working set size: 23998464.000000
estimated CPU time remaining: 45735.682079


After 7 min.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 40 C <-- TEMP. DROPS FAST !!


1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.846315
current CPU time: 24.267430 <-- AND NOW IT LOOKS LIKE NORMAL, COUNTING. BUT FIRST WHEN TEMP. DROPS ON GPU!?!?
fraction done: 0.017385
swap size: 17325019136.000000
working set size: 23998464.000000
estimated CPU time remaining: 45661.043715 <-- STILL LOOKS NORMAL, BUT THIS WILL STOP DECREASING AND START INCREAS LATER!?!?!?


After 9 min.:

# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 37 C <-- TEMP. DROPS FAST !!

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.846315
current CPU time: 179.339600 <-- AND NOW IT LOOKS LIKE NORMAL, COUNTING. BUT FIRST WHEN TEMP. DROPS ON GPU!?!?
fraction done: 0.017385
swap size: 17325019136.000000
working set size: 23998464.000000
estimated CPU time remaining: 45508.722678 <-- STILL LOOKS NORMAL, BUT THIS WILL STOP DECREASING AND START INCREAS LATER!?!?!?


After 10 min.:
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 35 C <-- TEMP. DROPS FAST !!

1) -----------
name: ps_140207_16179_23_1
WU name: ps_140207_16179_23
project URL: http://asteroidsathome.net/boinc/
report deadline: Thu Feb 27 23:31:38 2014
ready to report: no
got server ack: no
final CPU time: 0.000000
state: downloaded
scheduler state: scheduled
exit_status: 0
signal: 0
suspended via GUI: no
active_task_state: EXECUTING
app version num: 10111
checkpoint CPU time: 1.846315
current CPU time: 332.599800 <-- AND NOW IT LOOKS LIKE NORMAL, COUNTING. BUT FIRST WHEN TEMP. DROPS ON GPU!?!?
fraction done: 0.017385
swap size: 17325019136.000000
working set size: 23998464.000000
estimated CPU time remaining: 45357.533616 <-- STILL LOOKS NORMAL, BUT THIS WILL STOP DECREASING AND START INCREAS LATER!?!?!?


Installation Todo:
0. Take sudo rights
# sudo su

1. Make sure "gcc" is installed by checking version "gcc --version". Install "gcc"
# gcc --version
# apt-get install gcc

2. Reboot
# reboot

3. Download newest/best Linux Nvidia driver!? e.g."NVIDIA-Linux-x86_64-331.38.run" from nvidia.com
# wget http://us.download.nvidia.com/XFree86/Linux-x86_64/331.38/NVIDIA-Linux-x86_64-331.38.run
# chmod 0755 *.run
or
# chmod + x NVIDIA-Linux-x86_64-331.38.run

4. Blacklist nouveau and other nvidia modules. Edit "/etc/modprobe.d/blacklist.conf" and add "blacklist nouveau" etc. to the bottom.

# vi /etc/modprobe.d/blacklist.conf

blacklist vga16fb
blacklist nouveau
blacklist rivafb
blacklist nvidiafb
blacklist rivat

5. Make needed to install Nvidia driver
# apt-get install make

5. Reboot
# reboot

7. Take sudo rights
# sudo su

8. Install the downloaded Nvidia driver
# ./NVIDIA-Linux-x86_64-331.38.run

____________
Keep On Crunching!

Project RACK Linux Boinc Servers
Ubuntu Server 12.04.4 64bit
Intel i5/8Gb/P8H77-M
Asus GeForceGT640 2Gb
Nvidia v.nvidia-linux-x86_64-331.38
LM-Sensors v.3.3.1 with libsensors version 3.3.1
BOINC v.7.2.33 x86_64-pc-linux-gn

DanHansen@Denmark
Volunteer tester
Avatar
Send message
Joined: 14 Nov 12
Posts: 162
Credit: 2,056,165
RAC: 12,054
Denmark
Message 1478345 - Posted: 17 Feb 2014, 16:26:39 UTC
Last modified: 17 Feb 2014, 16:27:37 UTC

Hello friends ;)


Almost solved. Wrong edition, right result!

New attempt to solve the issue "Ubuntu Server 12.04.4 64bit - BOINC using GPU - Nvidia driver/GeForce GT610"

Just finishing the clean installation of Ubuntu Desktop 12.04.4 (Desktop, not Server!) Not being able to install the file from BOINC's download site, I did this instead "apt-get install boinc-client boinc-manager" which is showed on BOINC helping pages. But this does not install the newest boinc version. And now a notice tells me to download and install the new version.... etc. etc. never mind that right now.

But, after installing this desktop version, and installing boinc-client and boing-manager, and attaching to project using the manager, I checked to see if the GPU was running. Not yet, so I went on DL the newest driver from Nvidia 64bit 331.38. and making it bootable "chmod +x filename.run". Running the file gave me the same problems as when doing this on the Ubuntu Server Edition! So I went on, doing the same as I did there, almost.
I started a terminal, ctrl-alt-f2 and logged in, the claiming sudo rights "sudo su", then "stop lightdm", then running the Nvidia installer "sh NVIDIA-Linux-x86_64-331.38.run". Some questions are asked during the installation, I answered "Yes" to [install 32bit libraries etc.], "Yes" to [reconfigure X-server or something like that] (I will make a better todo when this is all finished!!)
Rebooting the system and updating the project from Boinc Manager showed me a 5 running unit. (CPU i5-3470 has 4 cores) and this 5 unit is named: (0.01 CPUs + 1 NVIDIA GPU). This I also found when doing it with Ubuntu Server 12.04.4, but the GPU ran for 7 minutes and the the temp. of the GPU suddenly dropped. This meaning the crunching also stopped. But not with the Ubuntu Desktop 12.04.4, here the GPU has been running now for 47 minutes and the GPU is still 47 degrees Celsius!
Here's the log file from Boinc Manager and a screen-dump of Boinc Manager showing the GPU crunching still after more than 10 minutes.:


GPU temp.:
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 44 C
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 44 C
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 45 C
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 45 C
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 45 C
# nvidia-smi -a |grep Gpu
Gpu : N/A
Gpu : 46 C


Mon 17 Feb 2014 04:07:24 PM CET | | Starting BOINC client version 7.0.27 for x86_64-pc-linux-gnu
Mon 17 Feb 2014 04:07:24 PM CET | | log flags: file_xfer, sched_ops, task
Mon 17 Feb 2014 04:07:24 PM CET | | Libraries: libcurl/7.22.0 OpenSSL/1.0.1 zlib/1.2.3.4 libidn/1.23 librtmp/2.3
Mon 17 Feb 2014 04:07:24 PM CET | | Data directory: /var/lib/boinc-client
Mon 17 Feb 2014 04:07:24 PM CET | | Processor: 4 GenuineIntel Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz [Family 6 Model 58 Stepping 9]
Mon 17 Feb 2014 04:07:24 PM CET | | Processor: 6.00 MB cache
Mon 17 Feb 2014 04:07:24 PM CET | | 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 rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase smep erms
Mon 17 Feb 2014 04:07:24 PM CET | | OS: Linux: 3.11.0-15-generic
Mon 17 Feb 2014 04:07:24 PM CET | | Memory: 3.81 GB physical, 3.95 GB virtual
Mon 17 Feb 2014 04:07:24 PM CET | | Disk: 54.68 GB total, 48.59 GB free
Mon 17 Feb 2014 04:07:24 PM CET | | Local time is UTC +1 hours
Mon 17 Feb 2014 04:07:24 PM CET | | NVIDIA GPU 0: GeForce GT 610 (driver version unknown, CUDA version 6.0, compute capability 2.1, 134215680MB, 134215667MB available, 134 GFLOPS peak)
Mon 17 Feb 2014 04:07:24 PM CET | | OpenCL: NVIDIA GPU 0: GeForce GT 610 (driver version 331.38, device version OpenCL 1.1 CUDA, 2048MB, 134215667MB available)
Mon 17 Feb 2014 04:07:24 PM CET | | Config: GUI RPC allowed from:
Mon 17 Feb 2014 04:07:24 PM CET | | A new version of BOINC is available. <a href=http://boinc.berkeley.edu/download.php>Download it.</a>
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | URL http://asteroidsathome.net/boinc/; Computer ID xxxxx; resource share 100
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | General prefs: from Asteroids@home (last modified 15-Oct-2013 16:31:32)
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | Computer location: work
Mon 17 Feb 2014 04:07:24 PM CET | | General prefs: using separate prefs for work
Mon 17 Feb 2014 04:07:24 PM CET | | Reading preferences override file
Mon 17 Feb 2014 04:07:24 PM CET | | Preferences:
Mon 17 Feb 2014 04:07:24 PM CET | | max memory usage when active: 1949.73MB
Mon 17 Feb 2014 04:07:24 PM CET | | max memory usage when idle: 3509.52MB
Mon 17 Feb 2014 04:07:24 PM CET | | max disk usage: 0.00GB
Mon 17 Feb 2014 04:07:24 PM CET | | (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
Mon 17 Feb 2014 04:07:24 PM CET | | Not using a proxy
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | Restarting task ps_140207_16280_78_1 using period_search version 10210 (sse3) in slot 0
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | Restarting task ps_140207_16280_32_0 using period_search version 10210 (sse2) in slot 1
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | Restarting task ps_140207_16280_73_0 using period_search version 10210 (sse2) in slot 2
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | Restarting task ps_140207_16280_22_1 using period_search version 10210 (sse2) in slot 3
Mon 17 Feb 2014 04:07:24 PM CET | Asteroids@home | Restarting task ps_140207_16088_26_2 using period_search version 10111 (cuda55) in slot 4
Mon 17 Feb 2014 04:08:03 PM CET | Asteroids@home | update requested by user
Mon 17 Feb 2014 04:08:05 PM CET | Asteroids@home | Sending scheduler request: Requested by user.
Mon 17 Feb 2014 04:08:05 PM CET | Asteroids@home | Requesting new tasks for CPU
Mon 17 Feb 2014 04:08:07 PM CET | Asteroids@home | Scheduler request completed: got 1 new tasks
Mon 17 Feb 2014 04:08:09 PM CET | Asteroids@home | Started download of input_16304_3
Mon 17 Feb 2014 04:08:11 PM CET | Asteroids@home | Finished download of input_16304_3


OK, I couldn't upload a screendump, it's not possible in this forum. The reason I would like to show you this screendump is because when 4 work units were done after about 50 minutes, the GPU had only completed about 5-6% of it's work unit.
I know this card GT610 is a "not very fast" card. We discussed this inhere. But I also learned that GPU's would be much faster doing data crunching that ordinary CPU's !! I will replace the GT610 with the faster Asus GeForce GT640 which I was told would be a much better choice. But how much better will it be?

What is the reason for doing all this work, if GPU's are so much "slower" than an ordinary Intel i5 CPU? How fast is a "fast" graphic card" compared to e.g. i5-3470?

We now know that the hardware in this test works, and we know that it runs on Ubuntu Desktop 12,04.4. OK, then why will it not run on Ubuntu Server 12.04.4? The only difference is the graphical environment, X-server! According to Ubuntu, there's not the difference of the Desktop Edition and the Server Edition as earlier on, e.g. Ubuntu Server 10.04 v. Ubuntu Desktop 10.04!
So why is this? Why will the GPU not work properly?

Any suggestions will be greatly appreciated ;)

.
____________
Keep On Crunching!

Project RACK Linux Boinc Servers
Ubuntu Server 12.04.4 64bit
Intel i5/8Gb/P8H77-M
Asus GeForceGT640 2Gb
Nvidia v.nvidia-linux-x86_64-331.38
LM-Sensors v.3.3.1 with libsensors version 3.3.1
BOINC v.7.2.33 x86_64-pc-linux-gn

DanHansen@Denmark
Volunteer tester
Avatar
Send message
Joined: 14 Nov 12
Posts: 162
Credit: 2,056,165
RAC: 12,054
Denmark
Message 1478731 - Posted: 18 Feb 2014, 13:06:40 UTC
Last modified: 18 Feb 2014, 13:09:00 UTC

Hello crunchers,


OK, I got the guys from Ubuntu Forums to join the fight!
And I found 2 very interesting guys from the Boinc Team at overclock.net too.

The solution with the Desktop Edition of Ubuntu 12.04.4 is just to complicated to handle, when you are used to the non-graphical environment of Ubuntu Server 12.04.4 !!

It looks like the Desktop Edition doesn't get as much data crunched as the Server Edition. And the Desktop is gets help from the GPU as well, but never the less it seems to be slower anyway!?!?

Look at this:
http://boincstats.com/en/stats/-1/user/detail/2560253/lastDays
Desktop Edition: --> 20.000 - 25.000 credits per day
Server Edition: ---> 25.000 - 35.000 credits per day

Well, we'll get the right numbers in a couple of days, but can this really be right?

From all the other forums out there which has a interest in Boinc or GPU data crunching in general, I discovered that many wish to be able to make a headless boinc cruncher.. Therefore I continue on this issue.

In the end I hope to be able to make a ToDo for all to use, for several different OS'. And to show how you can build headless RACK-mounted 2U Boinc Cruncher with no monitors attached, controlled an watched by shell scripts which alerts or even shuts down the server in question, should the CPU/GPU/HDD run hot. With a FAN controller with display (X-vision 1000), so that you can check the Boinc crunchers temp. in the RACK.
.
____________
Keep On Crunching!

Project RACK Linux Boinc Servers
Ubuntu Server 12.04.4 64bit
Intel i5/8Gb/P8H77-M
Asus GeForceGT640 2Gb
Nvidia v.nvidia-linux-x86_64-331.38
LM-Sensors v.3.3.1 with libsensors version 3.3.1
BOINC v.7.2.33 x86_64-pc-linux-gn

DanHansen@Denmark
Volunteer tester
Avatar
Send message
Joined: 14 Nov 12
Posts: 162
Credit: 2,056,165
RAC: 12,054
Denmark
Message 1488210 - Posted: 13 Mar 2014, 3:58:07 UTC
Last modified: 13 Mar 2014, 4:03:17 UTC

Hi,


I'm closing this thread down. No reason to continue here. Solved the case ;)

I will make a ToDo for those who are in the same situation as me. Wants to be part of e.g. SETI@home and have a few $ to spent. Want do make it a semi professional solution. The ToDo will be in one of the two links below. Along with a ToDO showing how to get a Rack, not so pricey and Cases as well. There's a little work to be done since the industrial 2U PSU's doesn't quite fit in the Cases. I'll show you how ;)

If you are in the same situation, you can follow the development here:

http://setiathome.berkeley.edu/forum_thread.php?id=73032
http://setiathome.berkeley.edu/forum_thread.php?id=74270

.
____________
Keep On Crunching!

Project RACK Linux Boinc Servers
Ubuntu Server 12.04.4 64bit
Intel i5/8Gb/P8H77-M
Asus GeForceGT640 2Gb
Nvidia v.nvidia-linux-x86_64-331.38
LM-Sensors v.3.3.1 with libsensors version 3.3.1
BOINC v.7.2.33 x86_64-pc-linux-gn

Questions and Answers : GPU applications : Ubuntu Server 12.04.4 - Use GPU to chrunch data

Copyright © 2014 University of California