Questions and Answers : Wish list : Machines vs Processors
Message board moderation
Author | Message |
---|---|
Send message Joined: 5 Aug 04 Posts: 55 Credit: 87,392 RAC: 0 |
Just noticed the new \'number of PCs\' column in the team stats which is nice, but would it not be more useful to have total number of processors instead. A 2 processor machine is pretty much 2 machines in one box by my reckoning. <a href="http://www.users.globalnet.co.uk/~sykesm/cpdn.html"><img src="http://www.users.globalnet.co.uk/~sykesm/gfx/sig.jpg"></a> |
Send message Joined: 5 Aug 04 Posts: 426 Credit: 2,426,069 RAC: 0 |
The problem with that is what do hyperthreaded machines count as? It isn't really 2 processors even though BOINC will recognize it as 2. <br>John Keck -- BOINCing since 2002/12/08 -- <a href="http://www.boinc.dk/index.php?page=user_statistics&project=cpdn&userid=191"><img border="0" height="80" src="http://191.cpdn.sig.boinc.dk?188"></a> |
Send message Joined: 5 Aug 04 Posts: 907 Credit: 299,864 RAC: 0 |
well sort of but then as jkeck notes what about HT's, and just because someone has 2 or 4 or 8 procs doesn't mean every single one is running CPDN etc. So I left it as just "one box!" |
Send message Joined: 5 Aug 04 Posts: 55 Credit: 87,392 RAC: 0 |
> well sort of but then as jkeck notes what about HT's, and just because someone > has 2 or 4 or 8 procs doesn't mean every single one is running CPDN etc. So I > left it as just "one box!" > Would it be easy to have a 'number of models' column instead then? |
Send message Joined: 5 Aug 04 Posts: 1283 Credit: 15,824,334 RAC: 0 |
> Would it be easy to have a 'number of models' column instead then? I'm not sure how accurate even that would be, Martin. If you count the number of uncompleted models on the system you'll include ones that have been downloaded but haven't started running plus any that have errored out without reporting back to the servers (at least until the server decides they're not trickling any more and can be reallocated). If you count the number of trickling models you'll again include errored out models and have a slight complication when a model completes. Do you assume another model has been started and use the same count (in which case lots of checks will be needed on the server) or do you reduce the count by one until the first trickle from the new model? <a href="http://www.teampicard.net"><img src="http://www.teampicard.net/templates/fisubice/images/phpbb2_logo.jpg"></a><a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/team_display.php?teamid=3">Join us here</a> |
©2025 cpdn.org