climateprediction.net (CPDN) home page
Thread 'takeover'

Thread 'takeover'

Message boards : Number crunching : takeover
Message board moderation

To post messages, you must log in.

AuthorMessage
old_user662116

Send message
Joined: 10 Sep 11
Posts: 2
Credit: 99,229
RAC: 0
Message 42876 - Posted: 11 Sep 2011, 2:18:59 UTC

I just started with this project, and the thing has taken over, my Seti has gone from 4 cores to 1 and I'm thinking that the Predict program runs for 400 hours as being a bit funny, and my Seti goes into hold mode, with half finished calcs.
I guess it's normal though, after all it's a Pom thing. VBG.(very BIG grin)
MaxG.
ID: 42876 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 42877 - Posted: 11 Sep 2011, 3:01:00 UTC - in response to Message 42876.  

Actually it's a BOINC thing.
You've just frightened the life out of your BOINC manager, by starting a project with VERY long run times, as compared to all of the other projects.
But the time estimate will go down a bit as BOINC 'learns' about these models, and eventually it will drop out of panic mode and let other work run as well.

And this IS climate modelling, NOT weather modelling.
Look at the graphics display, and you'll see that the models are due to run for 40 years. IF they don't become unstable and fail, due to 'silly' things like an old favourite, 'negative pressure'.

There's a description of the current project, RAPID-RAPIT, a Coupled Ocean model, here and here

Also, there are several threads in the Number crunching section of this board about problems, as well as the News and Announcements thread at the top of Number crunching, and another near the top of our alternative board, which is here.


Backups: Here
ID: 42877 · Report as offensive     Reply Quote
old_user662116

Send message
Joined: 10 Sep 11
Posts: 2
Credit: 99,229
RAC: 0
Message 42881 - Posted: 11 Sep 2011, 16:22:28 UTC

Well I don't mind if the "bosses" don't but I will not be around in 40 years so will it be wasted in this computer, or can I "will" it to some other guy with longivity beyond 2 years. At 81.5 old. that 40 years is way out of reach.
MaxG
ID: 42881 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 42883 - Posted: 11 Sep 2011, 20:58:10 UTC - in response to Message 42881.  

40 years "model time".
Did you look at the graphics display? You can see how fast it's actually running.

On my computers, 4 take about 3.5 weeks of calendar time.


Backups: Here
ID: 42883 · Report as offensive     Reply Quote
wateroakley

Send message
Joined: 6 Aug 04
Posts: 195
Credit: 28,337,510
RAC: 10,436
Message 42889 - Posted: 12 Sep 2011, 10:06:41 UTC - in response to Message 42881.  

Well I don't mind if the "bosses" don't but I will not be around in 40 years so will it be wasted in this computer, or can I "will" it to some other guy with longivity beyond 2 years. At 81.5 old. that 40 years is way out of reach.
MaxG

Welcome.
Looking at your first CM3 task 13373090 it has started running at 2.32 s/TS. If it's running 24/7, that extrapolates to around 4 elapsed weeks to complete the 40 model years.

You don't need to "will" it, as your WU hadcm3n_t0ft_1980_40_007453138 has already been "willed" to computer host 483262 as another task 13373091.
ID: 42889 · Report as offensive     Reply Quote

Message boards : Number crunching : takeover

©2024 cpdn.org