Currently my computer only runs one project at a time, totally negating the reason to load multiple projects. what is the setting that i\'ve missed out for them to run simultneously?

---

When you say they\'re running one at a time, are they taking turns at one hour intervals, or is it behaving differently?

You have a single-processor system, so if that\'s so, it\'s the expected behaviour. If you want to run two things simultaneously you\'d need to get a dual-core CPU.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

When you say they\'re running one at a time, are they taking turns at one hour intervals, or is it behaving differently?

On my machine (single CPU) it run constantly (as expected) but doesn\'t take any turn(s) at all - the second project (set on 1/10 of resource) keeps running all day long ... :-(

--
Arlekin\'s Dream Ltd. / Bernd Villiger et al.

It could be that the deadline on the other project is very close, so that Boinc may have decided it needs to concentrate on it for a while in order to make the deadline (this would show up on the messages tab as some gibberish about \'EDF scheduling\').

Alternatively, it could be a \'debt\' issue where CPDN had previously had more than it\'s fair share of CPU time, and so the other project is being given a boost to make up for that.

If either guess is right, in the long term it\'ll balance out, but in the short term one project gets an advantage over the other.

If you cut-and-paste the contents of the messages tab here, that may give us enough information to work out which alternative is the most likely.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

It could be that the deadline on the other project is very close, so that Boinc may have decided it needs to concentrate on it for a while in order to make the deadline (this would show up on the messages tab as some gibberish about \'EDF scheduling\').
Alternatively, it could be a \'debt\' issue where CPDN had previously had more than it\'s fair share of CPU time, and so the other project is being given a boost to make up for that.
If you cut-and-paste the contents of the messages tab here, that may give us enough information to work out which alternative is the most likely.

The messages tab shows these entries:
{code]
Fre 07 Jul 2006 07:31:17 CEST|climateprediction.net|Deferring task hadcm3lbm_bthc_05320691_1
Fre 07 Jul 2006 07:31:17 CEST|SETI@home|Resuming task 18mr99ab.11394.5328.228398.3.190_2 using setiathome_enhanced version 512
Fre 07 Jul 2006 07:31:17 CEST||Using earliest-deadline-first scheduling because computer is overcommitted.
Fre 07 Jul 2006 07:31:17 CEST||Suspending work fetch because computer is overcommitted.
{/code}
The strange this is that I the only way to let CDPN take it\'s time is to completely suspend SETI - otherwise SETI keeps taking all of the time (even reloading new tasks if necessary to keep BOINC busy ...).
--
Arlekin\'s Dream Ltd. / Bernd Villiger et al.

The overcommitted messages explain why you\'re having problems. The Boinc manager thinks it has downloaded too much work to do it all within their deadlines, so it\'s panicing and concentrating on the nearest deadline.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

The overcommitted messages explain why you\'re having problems. The Boinc manager thinks it has downloaded too much work to do it all within their deadlines, so it\'s panicing and concentrating on the nearest deadline.

Looks like it - since I cancelled some of the SETI tasks, CPDN is getting it\'s share of the cpu time (eager to know what will happen after the next update ...).

Thanks for helping.

---

To reinforce what Mike said, you have to take the long term view when running more than one project.
See what it does over 2-3 months, not over a couple of days.

And fiddling with the settings, (also called \'micro managing\'), will make it worse, as BOINC will sooner or latter do the same thing when you allow the other project to start. But now the time owned is greater.

To reinforce what Mike said, you have to take the long term view when running more than one project. See what it does over 2-3 months, not over a couple of days. (...) And fiddling with the settings will make it worse (...)

Message received and understood ;-) I\'ll try to be patient and keep my hand off the management software ...

Again: thanks for helping (and for being frank in the advice given).

--
Arlekin\'s Dream Ltd. / Bernd Villiger et al.

I am not getting credit for this machine. I run 3 machines and
get credit for 2. How can my id - brleed - get credit for
3 machines, not 2.

---

Your account only shows 2 machines running:
Computers

You did have this computer running: GenuineIntel Pentium III (Coppermine), but it hasn\'t been heard from since it crashed a model in March. (It\'s far too slow for this project anyway.)

If you have another computer, perhaps you put it on to another account.

There is one more possibility - perhaps that \'coppermine\' PC can\'t talk to the servers due to firewall or proxy issues. If you turn to the \'tasks/work\' tab, do you see anything listed, and shown as \'running\'? Also, if you click \'update\' on the first tab, then turn to the \'messages\' tab, are any error messages shown?

If you click on \'your account\' from the first tab (projects) on that PC, which account is displayed?

Incidentally the http://www.apsathome.org/ project runs local-scale atmospheric physics models (to be used to improve global climate models), and can run on just about any PC no matter how old.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

Of course, the \'slab\' model uses a lot less CPU power and memory, this option is available via Project Preferences / HadSM3 -> Ticked (untick HadCM3).

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ