BOINC not switching out of climateprediction

Questions and Answers : Windows : BOINC not switching out of climateprediction
Message board moderation

To post messages, you must log in.

AuthorMessage
squeak

Send message
Joined: 20 Jul 99
Posts: 27
Credit: 255,658
RAC: 0
Australia
Message 522045 - Posted: 23 Feb 2007, 7:17:15 UTC

I have set@home and climateprediction running, prefs say 60 minutes appl switch, 25% climate 75% seti@home but once climateprediction gets control it never relinquishes. Also climate prediction has a reduced Projects menu - no Preferences button. I am running 5.8.11
ID: 522045 · Report as offensive
Profile mikey
Volunteer tester
Avatar

Send message
Joined: 17 Dec 99
Posts: 4215
Credit: 3,474,603
RAC: 0
United States
Message 522085 - Posted: 23 Feb 2007, 10:49:07 UTC - in response to Message 522045.  

I have set@home and climateprediction running, prefs say 60 minutes appl switch, 25% climate 75% seti@home but once climateprediction gets control it never relinquishes. Also climate prediction has a reduced Projects menu - no Preferences button. I am running 5.8.11

CP has a time frame of a year in some cases, this means that your computer needs to get busy crunching or it will not be able to get them done on time. This will lead to what is called LTD or Long Term Debt. It will even out in the end, the end just being a while down the road.
ID: 522085 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 522498 - Posted: 24 Feb 2007, 4:03:37 UTC

Actually, since he is running 5.8.8, there is another distinct possibility. That may be that BOINC is going to wait for a checkpoint before switching - and CPDN does not checkpoint very often. But it is also true that CPDN may be running into deadline trouble, and may need extra CPU time to complete on time. Yes, I know that the deadline for CPDN is a bit on the loose side, but that is not true of most projects, and the BOINC daemon does not have any way of distinguishing whether the deadlines are strict for that project or not.

With the information given, I cannot tell which case is happening.


BOINC WIKI
ID: 522498 · Report as offensive
squeak

Send message
Joined: 20 Jul 99
Posts: 27
Credit: 255,658
RAC: 0
Australia
Message 522542 - Posted: 24 Feb 2007, 7:39:13 UTC

I'm running 5.8.11 as I said in the initial post, and the CP deadline is 4/2/2008, so I doubt that's a factor.
ID: 522542 · Report as offensive
Profile Darren

Send message
Joined: 18 May 99
Posts: 12
Credit: 1,642,195
RAC: 0
United Kingdom
Message 522634 - Posted: 24 Feb 2007, 12:43:19 UTC

I have exactly the same problem. I haven't done any SETI work for months.
ID: 522634 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14650
Credit: 200,643,578
RAC: 874
United Kingdom
Message 522706 - Posted: 24 Feb 2007, 16:38:03 UTC

The current CPDN 'coupled' models tell BOINC to do 1 full WU - that's 160 model years - in just under one calendar year. That would earn you 41,472 credits in 348 days, or just about 120 credits per day (taking the figures from my current CPDN WUs).

If you haven't got a machine which can sustain a steady RAC of 120 or more on CPDN, BOINC will always think you're in deadline trouble. Darren's computer has a RAC of 48, and squeak's is around 50, so it doesn't surprise me that neither machine is switching projects.

CPDN doesn't enforce the deadlines, and will accept whatever work you can do, but BOINC doesn't know that. I suggest that both of you decide which project is of most interest to you, and concentrate your energies on that one. By all means attach to a second project as a backup, because there are times when no work is available from the main project - but I would suggest that CPDN isn't really suitable as a backup.

ID: 522706 · Report as offensive
Profile Darren

Send message
Joined: 18 May 99
Posts: 12
Credit: 1,642,195
RAC: 0
United Kingdom
Message 522779 - Posted: 24 Feb 2007, 19:46:40 UTC - in response to Message 522706.  
Last modified: 24 Feb 2007, 19:57:27 UTC

The current CPDN 'coupled' models tell BOINC to do 1 full WU - that's 160 model years - in just under one calendar year. That would earn you 41,472 credits in 348 days, or just about 120 credits per day (taking the figures from my current CPDN WUs).

If you haven't got a machine which can sustain a steady RAC of 120 or more on CPDN, BOINC will always think you're in deadline trouble. Darren's computer has a RAC of 48, and squeak's is around 50, so it doesn't surprise me that neither machine is switching projects.

CPDN doesn't enforce the deadlines, and will accept whatever work you can do, but BOINC doesn't know that. I suggest that both of you decide which project is of most interest to you, and concentrate your energies on that one. By all means attach to a second project as a backup, because there are times when no work is available from the main project - but I would suggest that CPDN isn't really suitable as a backup.



Thanks for that. At least I know now.

I'm (hopefully) building a new system next week based on the Intel E6300 CPU. I'm hoping I'll go above 120 with that?
ID: 522779 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14650
Credit: 200,643,578
RAC: 874
United Kingdom
Message 522791 - Posted: 24 Feb 2007, 20:33:05 UTC - in response to Message 522779.  

I'm (hopefully) building a new system next week based on the Intel E6300 CPU. I'm hoping I'll go above 120 with that?

Should have absolutely no trouble at all, provided you have it switched on a reasonable proportion of the time, and don't spend too long looking at the pretty pictures (a.k.a. screensaver).
ID: 522791 · Report as offensive
Profile Darren

Send message
Joined: 18 May 99
Posts: 12
Credit: 1,642,195
RAC: 0
United Kingdom
Message 525027 - Posted: 1 Mar 2007, 18:10:53 UTC

I've built my new system now and it appears to be doing climateprediction and seti at the same time. Is it doing one in each core?
ID: 525027 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 525267 - Posted: 2 Mar 2007, 5:24:36 UTC - in response to Message 525027.  

I've built my new system now and it appears to be doing climateprediction and seti at the same time. Is it doing one in each core?

If it is a dual core machine, then yes, this is quite possible (don't be surprised if occasionally the machine runs two of one of the projects and none of the other for a while).


BOINC WIKI
ID: 525267 · Report as offensive
Profile Darren

Send message
Joined: 18 May 99
Posts: 12
Credit: 1,642,195
RAC: 0
United Kingdom
Message 525952 - Posted: 3 Mar 2007, 20:47:12 UTC

Well, after 3,408,813.34 seconds of CPU time the climate prediction client gave an error after crunching 20% of it. I'm fed up because it was doing this since August.

Don't think I'm going to bother with that project any more. I think I've only managed to finish one unit in all the time on it. Time to look around for another project that is worthy for my dual core to crunch on.
ID: 525952 · Report as offensive
Profile Darren

Send message
Joined: 18 May 99
Posts: 12
Credit: 1,642,195
RAC: 0
United Kingdom
Message 526227 - Posted: 4 Mar 2007, 9:34:28 UTC
Last modified: 4 Mar 2007, 9:35:59 UTC

In regards to my post above; both the seti unit I was working on and the cpdn model gave errors.

see:

http://setiathome.berkeley.edu/result.php?resultid=488249762

and

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5695358

I think this might have happened after I rebooted after I installed some drivers for my new mobo.

Any ideas what caused this?
ID: 526227 · Report as offensive
Profile mikey
Volunteer tester
Avatar

Send message
Joined: 17 Dec 99
Posts: 4215
Credit: 3,474,603
RAC: 0
United States
Message 526281 - Posted: 4 Mar 2007, 13:12:40 UTC - in response to Message 526227.  

In regards to my post above; both the seti unit I was working on and the cpdn model gave errors.
see:
http://setiathome.berkeley.edu/result.php?resultid=488249762
and
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5695358
I think this might have happened after I rebooted after I installed some drivers for my new mobo. Any ideas what caused this?


On the Seti unit this message came up
"<![CDATA[
<message>
The system cannot find the path specified. (0x3) - exit code 3 (0x3)"

Seems like you either shut down without stopping the unit and it crashed or you did more than just load new drivers.
The climate unit has the same error.
ID: 526281 · Report as offensive
Profile Darren

Send message
Joined: 18 May 99
Posts: 12
Credit: 1,642,195
RAC: 0
United Kingdom
Message 526333 - Posted: 4 Mar 2007, 15:58:37 UTC - in response to Message 526281.  

In regards to my post above; both the seti unit I was working on and the cpdn model gave errors.
see:
http://setiathome.berkeley.edu/result.php?resultid=488249762
and
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5695358
I think this might have happened after I rebooted after I installed some drivers for my new mobo. Any ideas what caused this?


On the Seti unit this message came up
"<![CDATA[
<message>
The system cannot find the path specified. (0x3) - exit code 3 (0x3)"

Seems like you either shut down without stopping the unit and it crashed or you did more than just load new drivers.
The climate unit has the same error.


I never stop BOINC before shutting down. Didn't know you were supposed to. Anyway, that cpdn model is duff now. I wasted a lot of CPU time on it.

ID: 526333 · Report as offensive
Aurora Borealis
Volunteer tester
Avatar

Send message
Joined: 14 Jan 01
Posts: 3075
Credit: 5,631,463
RAC: 0
Canada
Message 526367 - Posted: 4 Mar 2007, 17:22:42 UTC - in response to Message 526333.  
Last modified: 4 Mar 2007, 17:34:48 UTC

In regards to my post above; both the seti unit I was working on and the cpdn model gave errors.
see:
http://setiathome.berkeley.edu/result.php?resultid=488249762
and
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5695358
I think this might have happened after I rebooted after I installed some drivers for my new mobo. Any ideas what caused this?


On the Seti unit this message came up
"<![CDATA[
<message>
The system cannot find the path specified. (0x3) - exit code 3 (0x3)"

Seems like you either shut down without stopping the unit and it crashed or you did more than just load new drivers.
The climate unit has the same error.


I never stop BOINC before shutting down. Didn't know you were supposed to. Anyway, that cpdn model is duff now. I wasted a lot of CPU time on it.

It's should not be necessary to shut down Boinc before resetting your computer. This would seem to be just some random glitch probably associated with the update of the driver. I wouldn't worry about it unless it occurs again.

Your CPDN crunching in not wasted. The trickle reporting at CPDN assures that what data you did provide will be useful and that you get the credit for the work you did. Because for their extremely long crunch time and the odds of something going wrong over months of crunching, I've come to the conclusion that a relatively small number of people actually complete a full WU for that project, unless they devote all their computer resources to it. I've yet to finish one myself, but I still crunch for the project on my faster computer which is only slightly faster than yours. My CPDN is at 29% with 1400 hr to go. This is the first one on my C2D and I'm still hoping to finally complete a WU for this project. As you can see from my signature I crunch for a lot of projects and feel I contribute in a small way to all of them.

Boinc V7.2.42
Win7 i5 3.33G 4GB, GTX470
ID: 526367 · Report as offensive

Questions and Answers : Windows : BOINC not switching out of climateprediction


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.