climateprediction.net (CPDN) home page
Thread 'CP-boinc model run times.'

Thread 'CP-boinc model run times.'

Message boards : Number crunching : CP-boinc model run times.
Message board moderation

To post messages, you must log in.

AuthorMessage
Profileold_user156
Avatar

Send message
Joined: 5 Aug 04
Posts: 186
Credit: 1,612,182
RAC: 0
Message 6563 - Posted: 5 Dec 2004, 4:53:06 UTC
Last modified: 5 Dec 2004, 5:00:57 UTC

Looks like the difference between minimum and maximum model run times for CP-boinc is a little less than with classic CPDN;

<img src="http://cpdn.tuxie.org/uk_nick/Model_run_times.png">

I was regularly seeing up to 10% difference in run times on the same machine running classic CPDN but the maximum I've seen to date for CP-boinc is only 5% - ie. +/-2.5% of the average CPU time. (Computer specs via my sig.)

'Emma', my quickest machine, has only run 6.5 CP-boinc models as yet, so that difference may yet get somewhat larger over the longer term...

<a href="http://www.nmvs.dsl.pipex.com/"><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=6&amp;team=off&amp;trans=off"></a>

<a href="http://www.nmvs.dsl.pipex.com/">Distributed Mania</a>
ID: 6563 · Report as offensive     Reply Quote
Helmer Bryd

Send message
Joined: 16 Aug 04
Posts: 156
Credit: 9,035,872
RAC: 2,928
Message 7537 - Posted: 24 Jan 2005, 15:15:23 UTC - in response to Message 6563.  

Interesting. Do you have more data now?

ID: 7537 · Report as offensive     Reply Quote
Profileold_user156
Avatar

Send message
Joined: 5 Aug 04
Posts: 186
Credit: 1,612,182
RAC: 0
Message 7539 - Posted: 24 Jan 2005, 15:37:56 UTC

Yeah, I've finished a bunch more models but I'm very busy assembling new machines, modifying my current 'harem' and rebuilding 'oldsters' as hand-me-downs for my younger relatives - I'll get around to posting more up-to-date data eventually. ;-)

<a href="http://www.nmvs.dsl.pipex.com/"><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=6&amp;team=off&amp;trans=off"></a>

<a href="http://www.nmvs.dsl.pipex.com/">Distributed Mania</a>
ID: 7539 · Report as offensive     Reply Quote
Profileold_user17525

Send message
Joined: 13 Sep 04
Posts: 161
Credit: 284,548
RAC: 0
Message 7542 - Posted: 24 Jan 2005, 15:46:02 UTC
Last modified: 24 Jan 2005, 15:46:51 UTC

Well, I hope you're keeping a record of where you're up to with it all. When the php board goes back up it'll be the longest post. Ever ;-)

Marj
_________________________________
ID: 7542 · Report as offensive     Reply Quote
[B^S] mavau

Send message
Joined: 30 Aug 04
Posts: 142
Credit: 9,936,132
RAC: 0
Message 7552 - Posted: 24 Jan 2005, 18:12:35 UTC
Last modified: 24 Jan 2005, 18:17:08 UTC

My results show some <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/results.php?userid=3230">spread.</a>
The difference appears at startup. My latest completed model was fastest.
It shows an <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?field=Temperature&amp;resultid=430707&amp;phase=AT#graph">uncommon graph</a>
Current model is pretty slow.

Forum search Site search
ID: 7552 · Report as offensive     Reply Quote
Profileold_user156
Avatar

Send message
Joined: 5 Aug 04
Posts: 186
Credit: 1,612,182
RAC: 0
Message 7642 - Posted: 26 Jan 2005, 15:52:33 UTC
Last modified: 26 Jan 2005, 15:52:44 UTC

Cool models run faster, I have no idea why...

<a href="http://www.nmvs.dsl.pipex.com/"><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=6&amp;team=off&amp;trans=off"></a>

<a href="http://www.nmvs.dsl.pipex.com/">Distributed Mania</a>
ID: 7642 · Report as offensive     Reply Quote

Message boards : Number crunching : CP-boinc model run times.

©2024 cpdn.org