Message boards : Number crunching : WUs being rejected? Very short processing time.
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile jay

Send message
Joined: 10 Aug 07
Posts: 54
Credit: 813,704
RAC: 116
Message 19645 - Posted: 21 May 2008, 18:01:51 UTC

Not sure what is going on with a few WUs.
I didn't see other post about this - maybe didn't look in the right place... :-)

Wanted to share - Other crunchers having same results with WU- finishing WU in less than 10 seconds...

Here is BOINC Message:
5/21/2008 2:11:15 AM|lhcathome|Sending scheduler request: To fetch work. Requesting 17280 seconds of work, reporting 0 completed tasks
5/21/2008 2:11:20 AM|lhcathome|Scheduler request succeeded: got 1 new tasks
5/21/2008 2:11:22 AM|lhcathome|Started download of wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050.zip
5/21/2008 2:11:25 AM|lhcathome|Finished download of wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050.zip
5/21/2008 2:11:25 AM|Spinhenge@home|Sending scheduler request: To fetch work. Requesting 8640 seconds of work, reporting 0 completed tasks
5/21/2008 2:11:27 AM|lhcathome|Starting wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050_1
5/21/2008 2:11:27 AM|lhcathome|Starting task wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050_1 using sixtrack version 467
5/21/2008 2:11:29 AM|lhcathome|Computation for task wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050_1 finished
5/21/2008 2:11:31 AM|lhcathome|Started upload of wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050_1_0
5/21/2008 2:11:31 AM|Spinhenge@home|Scheduler request succeeded: got 2 new tasks
5/21/2008 2:11:34 AM|lhcathome|Finished upload of wm72z_m72z__33__64.31_59.32__10_12__6__10_1_sixvf_boinc361050_1_0


---

Looks like other crunchers finish the WU in less than 10 seconds..
see
http://lhcathome.cern.ch/lhcathome/workunit.php?wuid=2663896
also
http://lhcathome.cern.ch/lhcathome/workunit.php?wuid=2664436

----
If debug is needed, Please tell me what xml data to add.

Thanks for all your hard work!!!!
Jay
ID: 19645 · Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 2 Sep 04
Posts: 545
Credit: 148,912
RAC: 0
Message 19646 - Posted: 21 May 2008, 18:05:49 UTC

Tasks in LHC have an extrodinarily variable completion time. Some run fast like yours... the beam hit the wall, and some run for the 10,000; 100,000; or 1,000,000 turns ... taking hours ...
ID: 19646 · Report as offensive     Reply Quote
EclipseHA

Send message
Joined: 18 Sep 04
Posts: 47
Credit: 1,886,234
RAC: 0
Message 19649 - Posted: 22 May 2008, 2:57:39 UTC
Last modified: 22 May 2008, 3:00:40 UTC

Jay...

Long time crunchers have seen many of these "hit the wall" WU's, and it's far from uncommon. It's part of the science too....

The thing to note, is if the requested credit is something like 0.00xxxx, they'll stay marked as "pending' forever... I haven't counted for awhile, but I have many more than 100!

Don't bother asking about ridding yourself of these long term pending... Many folks have been asking the sysops for a lonnnnnnnnnnnng time...
ID: 19649 · Report as offensive     Reply Quote
Kaal

Send message
Joined: 7 Nov 05
Posts: 19
Credit: 248,179
RAC: 0
Message 19652 - Posted: 22 May 2008, 12:21:53 UTC

This question comes up quite frequently (actually both Qs: "Why so short...?" & "What's happening with the long term pending...?")
Perhaps a sticky from N&A with the answers?
Q: "Why did my WU only last mere seconds?"
A: Because the 'beam' hit the 'wall' in the simulation. Don't worry it's still valid science which they need.
Q: "I've got some results with 0.0000...{insert tiny number} credit which have been hanging around for ages. What's up with that?"
A: Those results are from the short 'hit the wall' runs and (as I understand it) a server-side upgrade of the BOINC software is needed before a comprehensive clear-out can happen.
Q: "Will the server-side software upgrade happen any time soon?"
A: Haaaaaahahahahahaha!
ID: 19652 · Report as offensive     Reply Quote
KAMasud

Send message
Joined: 7 Oct 06
Posts: 114
Credit: 23,192
RAC: 0
Message 19654 - Posted: 23 May 2008, 6:30:00 UTC


I am just wondering as to how much time we waste on these wall crashers taking (Forget the science for the time being)into consideration that no credit is granted. Now i have four pages of them, all pending.
Regards
Masud.
ID: 19654 · Report as offensive     Reply Quote

Message boards : Number crunching : WUs being rejected? Very short processing time.


©2024 CERN