Very OT -- Subscribing to the Climate Prediction program.

Quoted message said:

we don't contribute to global warming other than by our normal bodily
functions

Unfortunately running number crunching programs when you PC is idle is a
great way to contribute to global warming.

From http://www.pcpro.co.uk/features/71029/electric-shock/page2.html

The amount of power required by a modern processor is proportional to how
hard it is working. And we are not talking about a couple of watts here
either. We tested a 3GHz Pentium 4 system and found that its power
consumption while idle was around 110W. But then we started up SETI@home and
watched the consumption increase by an enormous 90W as the CPU, chipset and
RAM subsystems swung into saturation from the intensity of the calculations.
So even if you leave your PC on and idle all the time - although that's bad
enough in terms of energy conservation - it is even worse if you set it to
churn through hefty calculations. This is coupled with the competitive
aspect that has emerged in grid computing, with teams of users competing to
'win' and be at the top of the computing league table. So not only has it
led to more energy being used in computers that would have been left on
anyway, but more computers have been left on that would otherwise have been
turned off in order to accumulate number-crunching points.

POHB said:

Quoted message said:

we don't contribute to global warming other than by our normal bodily
functions

Unfortunately running number crunching programs when you PC is idle is a
great way to contribute to global warming.

From http://www.pcpro.co.uk/features/71029/electric-shock/page2.html

<Informative stuff snipped>

Intersting. Any idea how the extra energy consumption resulting from
grid computing might compare with the energy consumption of tackling
the calculations with super-computers in the traditional way?

Additionally, if grid computing makes it feasible to work on a problem
that otherwise would or could not be tackled at all, can a value be
placed on the knowledge gained to be offset against the energy
consumption? Also, in the case of the Climate Prediction Program the
knowledge gained /might/ allow us to take actions to reduce global
warming that would offset the extra energy consumption millions of
times over.

--
Dave...

dkahn400 said:

Intersting. Any idea how the extra energy consumption resulting from
grid computing might compare with the energy consumption of tackling
the calculations with super-computers in the traditional way?

That's an easy question - the distributed spare-cycles grid way of
doing it will use an infinite amount more energy than doing it
on a traditional supercomputer since the calculation wouldn't
get done any other way!

Cheers,

Arthur

--
Arthur Clune

POHB said:

The amount of power required by a modern processor is proportional to how
hard it is working. And we are not talking about a couple of watts here
either. We tested a 3GHz Pentium 4 system and found that its power
consumption while idle was around 110W. But then we started up SETI@home and
watched the consumption increase by an enormous 90W as the CPU, chipset and
RAM subsystems swung into saturation from the intensity of the calculations.

Get a low-energy PC. Mine runs on about 40 watts, maxing out at about
60 when working at full whack. The monitor is another 45 when on.

--
not me guv

Nick Kew said:

POHB said:

The amount of power required by a modern processor is proportional
to how hard it is working. And we are not talking about a couple of
watts here either. We tested a 3GHz Pentium 4 system and found that
its power consumption while idle was around 110W. But then we
started up SETI@home and watched the consumption increase by an
enormous 90W as the CPU, chipset and RAM subsystems swung into
saturation from the intensity of the calculations.

Get a low-energy PC. Mine runs on about 40 watts, maxing out at about
60 when working at full whack. The monitor is another 45 when on.

Fair point. There's still a non negligible energy use per calculation,
though.

--
Ambrose

dkahn400 said:

Intersting. Any idea how the extra energy consumption resulting from grid
computing might compare with the energy consumption of tackling the
calculations with super-computers in the traditional way?

We did figures on performance per watt, to evaluate energy consumption of
different processors.
Presented at the Daresbury Machine Evaluation
Workshop two years ago
http://www.cse.clrc.ac.uk/disco/mew15-cd/Talks/Pearce_Streamline.pdf

Sad to say it, but 'traditional' supercomputers are pretty dead. Linux
clusters dominate the Top 500. Yours truly on the team which built no 129
last year.

Arthur Clune said:

That's an easy question - the distributed spare-cycles grid way of doing
it will use an infinite amount more energy than doing it on a traditional
supercomputer since the calculation wouldn't get done any other way!

Going even further off topic, I would not characterise folding at home,
SETI etc. schemes for using distributed computing at home users as grid
computing.

John Hearns said:

Going even further off topic, I would not characterise folding at home,
SETI etc. schemes for using distributed computing at home users as grid
computing.

That's part of the problem - everyone seems to have a different idea of
what grid computing is.

A bit like bicycling, really. Except for the (lack of) funding.

R.

John Hearns said:

Sad to say it, but 'traditional' supercomputers are pretty dead. Linux
clusters dominate the Top 500.

The Earth Simulator was top for quite a while, and I wouldn't bet
against its 10PFlop replacement (still on the drawing board) doing much
the same.

I'll not comment on how sensible this strategy is!

James
--
James Annan
see web pages for email
http://www.ne.jp/asahi/julesandjames/home/
http://julesandjames.blogspot.com/

I have just installed a new 600W PSU in this computer because my NVidia
6600GT kept crashing the machine when 3D acceleration kicked in.
More Co2.

"Pinky" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

I have just reopened access to the www.climateprediction.net organised by
Oxford University. I ran it for about a year until it changed its
procedures and my creaking pc didn't like it.
So today I have reopen my connection and while I was setting it up and
downloading updates I had this idea..............

If you already subscribe and take part in the BOINC climate prediction net
you will know what I am talking about. If not, it is a programme that (
like SETI@home) uses idle time in peoples running pcs to run a very long
programme which does comparisons of 3 time periods of weather prediction.

As a sort of side line there are teams to encourage members to chalk up
chuff (brownie)points. I was a member of a team when I first started doing
it a couple of years ago.

BUT

This time I thought -- I bet some of the lurkers on u.r.c already
subscribe and there may be others who would like to do so.

AND --- why not a "UK Cyclists" team.

SO -- I have set it up and there is a UK Cyclists team with 1 member ( me)
If you are already a subscriber then you might like to consider joining a
new team which is cycle aware/friendly

If anyone is interested do have a look at the website and it has the
vaguest connection with cycling -- in that we don't contribute to global
warming other than by our normal bodily functions. There is no need for a
great input if you decide to participate and I shall certainly not act the
part of a busybody moderator within the team. Just let it run and just
keep an eye on those who let it lapse so they can leave the team. A lot of
the teams do have "dead wood" members who have ceased to actively take
part.

The Climate prediction net does do better if your pc is on for long hours
every day. Some people who control office system also take part and they
run up huge numbers of brownie points -- but it would be nice to chug
along gently making a Cyclists team have its part to play

After downloading the control programme ( BOINC) you will see that there
are other programmes that can be run. I have run SETI ( Search for Extra
terrestrial Intelligence) for several years off and on and their programme
has a much more rapid turnover ( I have set up a "UK Cyclists" team in the
SETI programme as well!)

Anyway that is enough.

I don't want to start off a long OT topic on here but I thought some of
you might be interested. If you are don't respond on here but go to the
either
www.climateprediction.net or

http://setiathome.berkeley.edu./

and sign up!

It is the first day of Spring tomorrow. I am, at last, starting to clock
up some miles and feeling better day by day!

Trevor A Panther
In South Yorkshire,
England, United Kingdom.
www.tapan.pwp.blueyonder.co.uk

"ilikebikes" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

I have just installed a new 600W PSU in this computer because my NVidia
6600GT kept crashing the machine when 3D acceleration kicked in.
More Co2.

OTOH I've noticed a distinct correlation between people who cycle long
distances and who own *stacks* of top of the range equipment / build linux
clusters in their own bedrooms / run half the Internet from their garden
shed etc..

Perhaps cycling helps balance out the climate change left by all the
computers switched on? 🙂

Alex

Mr R@T / General Lighting
Ipswich, Suffolk, Untied Kingdom
http://www.partyvibe.com

On Tue, 21 Mar 2006 17:47:09 -0000, "Mr R@t \(2.30 zulu-india\)"
<[email hidden]> said in <[email hidden]>:

Quoted message said:

OTOH I've noticed a distinct correlation between people who cycle long
distances and who own *stacks* of top of the range equipment / build linux
clusters in their own bedrooms / run half the Internet from their garden
shed etc..

Tee-hee :-)

Guy
--
May contain traces of irony. Contents liable to settle after posting.
http://www.chapmancentral.co.uk

85% of helmet statistics are made up, 69% of them at CHS, Puget Sound