Planetarion Forums

Planetarion Forums (https://pirate.planetarion.com/index.php)
-   General Discussions (https://pirate.planetarion.com/forumdisplay.php?f=2)
-   -   BBC climate change (https://pirate.planetarion.com/showthread.php?t=190263)

DrNick 21 Mar 2006 02:53

BBC climate change
 
HI EVERYBODY!

Appologies for the loud call there but habit preceeds thought in that respect. Anyways I'm running the BBC climate change experiment. The basis of this experiment is to utilise the power of distributed computing, that is use loads of spare joe bloggs computer time to run cpu heavy scientific models, to predict climate change.

I've been running the model for nearly a week solid now on a dual core 4200+ and I'm 4.11% in. This worries me because the deadline for each submission is 1 year from now. After a few quick sums on calculator I figure it will take be roughly 1/3 of a year to complete my model. This is with me leaving my computer on 24/7 and I have a pretty high spec comp.

My problem is that the average person wont have their computer on every hour of every day, or even a third of that. It seems a bit of a fool-hardy experiment. Also the model seems quite random. In the year 1925 most of the northern hemisphere of my model was covered in ice and now in 1927 it seems like it's back to normal. I know their models will be subject to errors i'm not aware of but surely they will be itterating though a set of environmental factors they are unsure of. In this case what kind of models are they running if the world is surrounded in ice one year and blissfully warm the next.

My major problem is that i doubt more than 5% of users who download the program will run it to completion. They don't seems to have anticipated the level of dedication the average user will have in terms of running their models. If i leave my computer on for the next 5 months without using it I will finish my model. Like I said I'm running a dual core 4200+ and I assume that's above average.

I'm not sure what I hope to achieve by posting this but I just had to say that I'm beginning to feel like I'm running a public awareness experiment rather than an actual experiment.

Yahwe 21 Mar 2006 02:59

Re: BBC climate change
 
http://pirate.planetarion.com/forumdisplay.php?f=57

DrNick 21 Mar 2006 03:03

Re: BBC climate change
 
what's that supposed to mean? Isn't this a GD thread? The BBC is pretty general.

Ragnarak 21 Mar 2006 16:00

Re: BBC climate change
 
Strictly speaking it's not the BBC, it's the AOPP group at Oxford (amongst others).

I thought the models were meant to take a few weeks/months to run. Also I didn't think they supported dual core processors but would have to check that to be sure.

As for your ice worlds, that's the whole point of running these models isn't it?

Phang 21 Mar 2006 16:33

Re: BBC climate change
 
i turned mine off because it completed 0.0013% in three weeks and ate far too much memory while idle.

madi 21 Mar 2006 17:13

Re: BBC climate change
 
cant you link a few computers up to do it faster?
i am no geek but i have 7 or 8 lurking about in the spare room, ask around and find a few people who have spare or second machines they can lend you (tell them for 8 weeks or so, doesnt sound too long and gives them an end point which you can later extend) then set them all running together (dunno how you do this bit but i am sure someone does) and leave them whirring away for a few months while you still have your PC to potter about on.

Phil^ 21 Mar 2006 17:16

Re: BBC climate change
 
you mean as a beowulf cluster or similar? , technically yes but practically no, since the software is likely to be windows only

Akujin 21 Mar 2006 20:02

Re: BBC climate change
 
*spanks madi and runs aweh*

Phang 21 Mar 2006 20:07

Re: BBC climate change
 
Quote:

Originally Posted by Akujin
*spanks madi and runs aweh*

have you considered a career in public broadcasting?

Mushroom 21 Mar 2006 21:37

Re: BBC climate change
 
I find it vaguely amusing that they're attempting to study the effects of climate change by encouraging lots of people to leave their computers on processing large amounts of data and doubtless using rather a lot of energy in the process :p

Ragnarak 21 Mar 2006 22:33

Re: BBC climate change
 
Each day, about 23 times more energy will be spent boiling water for tea in the United Kingdom than would be used by the computers involved in the climateprediction.net project. More seriously, a rough calculation suggests that 100,000 computers running 24hrs/day for one year at a power consumption of 50W will contribute approximately 0.0001% of the total amount of CO2 generated in one year. This is not an insignificant amount, but seems (to us) a worthwhile investment to better understand the climate system.

Stew 21 Mar 2006 22:54

Re: BBC climate change
 
Quote:

Originally Posted by Ragnarak
(to us)

?

Ragnarak 21 Mar 2006 23:17

Re: BBC climate change
 
Quote:

Originally Posted by Stew
?

I was quoting from their site. Surprisingly enough their FAQ addresses most things people have commented on.

This is however fairly close to what I "do" now (in so much as I'm aware of people/work in the field). Considering the PI and a lot of people working on this are in Oxford's AOPP Group I assume Jenny probably knows quite a bit about it too.

Anyway, how's stuff? PM me or something.

Radical Edward 22 Mar 2006 11:03

Re: BBC climate change
 
Quote:

Originally Posted by Ragnarak
Each day, about 23 times more energy will be spent boiling water for tea in the United Kingdom than would be used by the computers involved in the climateprediction.net project. More seriously, a rough calculation suggests that 100,000 computers running 24hrs/day for one year at a power consumption of 50W will contribute approximately 0.0001% of the total amount of CO2 generated in one year. This is not an insignificant amount, but seems (to us) a worthwhile investment to better understand the climate system.


apparently even just leaving things in standby uses up a couple of powerstations worth of emissions every year.

Ste 22 Mar 2006 11:18

Re: BBC climate change
 
i'm an Environmental Coordinator at the moment too :)


All times are GMT +1. The time now is 20:26.

Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2002 - 2018