Advanced search

Message boards : Number crunching : My first NATHAN failed

Author Message
Profile Saenger
Avatar
Send message
Joined: 20 Jul 08
Posts: 134
Credit: 23,657,183
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwat
Message 22047 - Posted: 10 Sep 2011 | 14:04:15 UTC
Last modified: 10 Sep 2011 | 14:05:46 UTC

I1R7-NATHAN_FA2-1-100-RND1497_0
No wingman yet, hasn't been resend.

Run time 1.011565
CPU time 2.68
stderr out

<core_client_version>6.10.58</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
# Using device 0
# There is 1 device supporting CUDA
# Device 0: "GeForce GT 240"
# Clock rate: 1.34 GHz
# Total amount of global memory: 536150016 bytes
# Number of multiprocessors: 12
# Number of cores: 96
SIGABRT: abort called
Stack trace (13 frames):
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31(boinc_catch_signal+0x4d)[0x4819cd]
/lib/libc.so.6(+0x33af0)[0x7f40a2b22af0]
/lib/libc.so.6(gsignal+0x35)[0x7f40a2b22a75]
/lib/libc.so.6(abort+0x180)[0x7f40a2b265c0]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x48f4ab]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x4341dc]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x430cd6]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x4303e7]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x414d99]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x407b1a]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x4083fe]
/lib/libc.so.6(__libc_start_main+0xfd)[0x7f40a2b0dc4d]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x407899]

Exiting...

</stderr_txt>
]]>


Is it something with my computer or a fault of the WU?

Edith asks:
How could the run time be less than the CPU-time? Multi-CPU-WU?
____________
Gruesse vom Saenger

For questions about Boinc look in the BOINC-Wiki

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22054 - Posted: 11 Sep 2011 | 9:41:10 UTC - in response to Message 22047.

Someone else had no problems processing that WU.
http://www.gpugrid.net/workunit.php?wuid=2694264

This indicates that it is your machine that caused the issue, not the WU.

As for using more CPU than GPU ... in the first couple of seconds this will always be true as the WU needs *stuff* from the main computer which it has to talk through the CPU to get.
____________
Thanks - Steve

Profile Saenger
Avatar
Send message
Joined: 20 Jul 08
Posts: 134
Credit: 23,657,183
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwat
Message 22078 - Posted: 12 Sep 2011 | 17:09:43 UTC - in response to Message 22054.
Last modified: 12 Sep 2011 | 17:11:28 UTC

Someone else had no problems processing that WU.
http://www.gpugrid.net/workunit.php?wuid=2694264

This indicates that it is your machine that caused the issue, not the WU.

As for using more CPU than GPU ... in the first couple of seconds this will always be true as the WU needs *stuff* from the main computer which it has to talk through the CPU to get.


As for the CPU:
It's not using more CPU-time than GPU-time, but more CPU-time than real-time.

As for NATHAN's:
The second one failed: I9R20-NATHAN_FA2-4-100-RND8746_0
<core_client_version>6.10.58</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
# Using device 0
# There is 1 device supporting CUDA
# Device 0: "GeForce GT 240"
# Clock rate: 1.34 GHz
# Total amount of global memory: 536150016 bytes
# Number of multiprocessors: 12
# Number of cores: 96
SIGABRT: abort called
Stack trace (13 frames):
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31(boinc_catch_signal+0x4d)[0x4819cd]
/lib/libc.so.6(+0x33af0)[0x7fb6dcd96af0]
/lib/libc.so.6(gsignal+0x35)[0x7fb6dcd96a75]
/lib/libc.so.6(abort+0x180)[0x7fb6dcd9a5c0]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x48f4ab]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x4341dc]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x430cd6]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x4303e7]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x414d99]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x407b1a]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x4083fe]
/lib/libc.so.6(__libc_start_main+0xfd)[0x7fb6dcd81c4d]
../../projects/www.gpugrid.net/acemd2_6.14_x86_64-pc-linux-gnu__cuda31[0x407899]

Exiting...

</stderr_txt>
]]>


This time more time on the clock than used CPU ;)

If it's my computer, that crunches most other GPUgrid stuff flawless, what's it?
It's a Linux (ubuntu 10.04) Intel (C2D9450@3.2GHz) nVidia (GT240) with new drivers (280.13) and 8GB RAM running BOINC 6.10.58.

The other WU type that fails always is GPCR:
74-KASHIF_GPCR_14_ba1-8-100-RND2374_2
Do they have anything in common the others don't have?
____________
Gruesse vom Saenger

For questions about Boinc look in the BOINC-Wiki

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22081 - Posted: 12 Sep 2011 | 22:58:53 UTC - in response to Message 22078.

Might be a coincidence but your wingmen's tasks validated on the 6.15app (Win only), so perhaps these NATHAN tasks prefer 6.15.

Dagorath
Send message
Joined: 16 Mar 11
Posts: 509
Credit: 179,005,236
RAC: 0
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22084 - Posted: 12 Sep 2011 | 23:56:48 UTC - in response to Message 22081.

I had a NATHAN crash on my Linux box but it also crashed on a Win7 system. See here.

This NATHAN crashed on my Linux box, plus another Linux, plus a Win7.

This NATHAN validated on my Linux and I have had other NATHANs validate too so I don't know if we can conclude NATHANs prefer the 6.15 app. Maybe they're just crash prone?

Betting Slip
Send message
Joined: 5 Jan 09
Posts: 670
Credit: 2,498,095,550
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22444 - Posted: 1 Nov 2011 | 12:01:11 UTC - in response to Message 22084.

Even on a Windows 7 machine they use 98% GPU so they give card a real hammering and make GUI stutter. I am an hour away from completing one on my 460GT
____________
Radio Caroline, the world's most famous offshore pirate radio station.
Great music since April 1964. Support Radio Caroline Team -
Radio Caroline

Betting Slip
Send message
Joined: 5 Jan 09
Posts: 670
Credit: 2,498,095,550
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22445 - Posted: 1 Nov 2011 | 13:36:12 UTC - in response to Message 22444.
Last modified: 1 Nov 2011 | 13:39:52 UTC

Credits not very good either :( Just over 16,000 for over 36,000 secs including 50% bonus.
____________
Radio Caroline, the world's most famous offshore pirate radio station.
Great music since April 1964. Support Radio Caroline Team -
Radio Caroline

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 22448 - Posted: 1 Nov 2011 | 22:51:11 UTC - in response to Message 22445.
Last modified: 2 Nov 2011 | 8:43:08 UTC

There are goods and bads about Nathan's tasks (like all tasks).
The short tasks finish nice and quick on my systems, and never fail to get the full bonus (unlike other tasks). The long tasks on the other hand don't seem to give the best credit, but they are relatively shorter (and again always finish early enough for the full bonus). I expect extra credit is being given to 'other' tasks that require more CPU time. Nate's tasks don't use much CPU at all (another plus).
So far I don't think I have had any failures, but I can confirm that Nate's tasks are running at 99% on my overclocked (687MHz) GTX470, something I was not expecting! On my Apm GPU temps are 62deg C, about 5deg over other tasks, but so far no lag (another plus).
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Post to thread

Message boards : Number crunching : My first NATHAN failed

//