climateprediction.net (CPDN) home page
Thread 'sulphur_cycle deadline unreachable'

Thread 'sulphur_cycle deadline unreachable'

Questions and Answers : Unix/Linux : sulphur_cycle deadline unreachable
Message board moderation

To post messages, you must log in.

AuthorMessage
copycat

Send message
Joined: 24 Feb 05
Posts: 28
Credit: 121,749
RAC: 0
Message 15718 - Posted: 5 Sep 2005, 20:54:01 UTC

Running BOINC 4.43, with 3 projects: Seti@home, Einstein@home and, recently (re-)attached, CPDN. I tried CPDN before (some time ago) but it kept crashing so I abandoned it. The sulphur_cycle computations seem to be very cpu-intensive because the progress % is going up very slowly and the time to complete is enormous: 1463:06:30 as I write this, with a deadline of 2006/01/24. Given the fact my computer does not run 24/24, 7/7, far from it (15-20 hours/week at best, is my best guess), I calculated I can never reach the deadline. Although there was a \'no work fetch allowed\'-policy CPDN downloaded another WU, be it one for the hadsm3-application. That requires \'only\' 691:11:30 hours to completion, and has a deadline of 2006/08/16.
So, what do I do? Abort the sulphur_cycle now, and thus relieve my pc of this \'overcommitted\'-burden? Wait for the deadline, and abort the sulphur-WU then? Wait until it has finished the sulphur-WU, although that will finish several months AFTER its deadline and will possibly shorten the time I can work on the other CPDN-WU? Please advise. Thanks in advance.
ID: 15718 · Report as offensive     Reply Quote
Desti

Send message
Joined: 6 Aug 04
Posts: 124
Credit: 9,195,838
RAC: 0
Message 15720 - Posted: 5 Sep 2005, 23:04:27 UTC

15-20 hours/week isn't that much.
I would cancel it.

<br>
<a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/team_display.php?teamid=43"><img src="http://www.boincstats.com/stats/teambanner.php?id=6673"></img></a>
ID: 15720 · Report as offensive     Reply Quote
ProfileastroWX
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1496
Credit: 95,522,203
RAC: 0
Message 15721 - Posted: 5 Sep 2005, 23:04:28 UTC

CPDN deadlines are not "drop dead" times; in CPDN, results should be accepted after deadlines.

The short deadlines for SC were set because the results are required for start of the Coupled Model.

If you plan to abort the Model, no sense in waiting. If you continue to run it, thinking to abort it at the deadline, boinc will put everything else on hold trying to meet the SC deadline.

More info <a href="http://boinc-doc.net/boinc-wiki/index.php?title=Sulphur_Cycle">here.</a>

___________________________________________
ID: 15721 · Report as offensive     Reply Quote
copycat

Send message
Joined: 24 Feb 05
Posts: 28
Credit: 121,749
RAC: 0
Message 15752 - Posted: 6 Sep 2005, 21:10:54 UTC

I know everything is put on hold while BOINC is trying to meet the SC deadline. However, that does not bother me. Those are the only two WU's left, S@H has none left, E@H has none left. I am not planning to abort the model if the results will still be accepted (and credited) after the deadline. I was just wondering, since in E@H I've had WU's which were sent in some time after the deadline and thus did not get any credit. Anyway, if I abort this one, CPDN will probably deposit another one in my BOINC so I'll be right back where I started. I am not doing this for the credit, but I don't like my computer working so long and hard on something and then not get anything in return for it. So, if I get confirmation the SC-WU will still get validated/credited when it's sent back weeks/months after the deadline (and thus miss the start of the Coupled Model) I'll keep it running. If I don't get anything in return, I'll abort it. I know S@H and E@H will come back once these (CPDN-WU's) are finished. Might be in a few months, but so what?
ID: 15752 · Report as offensive     Reply Quote

Questions and Answers : Unix/Linux : sulphur_cycle deadline unreachable

©2024 cpdn.org