Welcome Guest [Log In] [Register]
We hope you enjoy your visit.


You're currently viewing our forum as a guest. This means you are limited to certain areas of the board and there are some features you can't use. If you join our community, you'll be able to access member-only sections, and use many member-only features such as customizing your profile, sending personal messages, and voting in polls. Registration is simple, fast, and completely free.


Join our community!


If you're already a member please log in to your account to access all of our features:

Username:   Password:
Add Reply
Global Warming Twice as Bad as Previously Thought
Topic Started: Jan 27 2005, 09:32 AM (514 Views)
Admiralbill_gomec
UberAdmiral
Wichita
Jan 27 2005, 01:11 PM
Make that three out of three.

If you think that he is baiting you, the response is easy. Don't respond.

If not one is interested, the topic drops off the page. If people are and can discuss things rationally, it stays.

I'm sorry, but I simply refuse to follow along with this hysteria like one of a flock of sheep. In addition, I refuse to propagate these lies. Spreading these lies (to me) equates to both flaming and baiting. That's MY take.

:no:



Offline | Profile | Quote | ^
 
Dandandat
Member Avatar
Time to put something here
Admiralbill_gomec
Jan 28 2005, 07:38 AM
Spreading these lies (to me) equates to both flaming and baiting. That's MY take.

To bad my friend, others peoples view points in and of them selves are not flaming or batting. Posting something that Admiralbill_gomec doesn’t agree with is not flaming or batting. Every time someone posts an article that has nothing to do with Admiralbill_gomec personally, and Admiralbill_gomec responds by saying the poster is flaming or batting, it is Admiralbill_gomec that is flaming and batting and it will not be tolerated.

Don’t agree with the article? Fine nether did I. All you have to do is respond to the article and explain in your often knowledgeable way why the article is wrong. There is no need to speak to or about the poster that you don’t like.
Offline | Profile | Quote | ^
 
captain_proton_au
Member Avatar
A Robot in Disguise

somerled
Jan 28 2005, 02:45 AM

This is a very good example of the application of massive parallelism , ie lots of CPUs all working at once , via TCP , on the tiny parts of a huge calculation problem , a poor man's SUPERCOMPUTER.


Well thats pretty much how the supercomputers nowadays are constructed, just 5000 cpu's patched together
Offline | Profile | Quote | ^
 
somerled
Member Avatar
Admiral MacDonald RN
Dwayne
Jan 28 2005, 07:33 AM
Quote:
 
Alarm at new climate warning

By Richard Black - BBC environment correspondent 

Temperatures around the world could rise by as much as 11C, according to one of the largest climate prediction projects ever run.

This figure is twice the level that previous studies have suggested.

Scientists behind the project, called climateprediction.net, say it shows that a "safe" upper limit for carbon dioxide is impossible to define.

The results of the study, which used PCs around the world to produce data, are published in the journal Nature.

Climateprediction.net is run from Oxford University, and is a distributed computing project; rather than using a supercomputer to run climate models, people can download software to their own PCs, which run the programs during downtime.

More than 95,000 people have registered, from more than 150 countries; their PCs have between them run more than 60,000 simulations of future climate.

Each PC runs a slightly different computer simulation examining what happens to the global climate if levels of carbon dioxide in the atmosphere double from pre-industrial levels - which may happen by the middle of the century. [Stress the phrase 'may happen']

What vary most between the simulations are the precise nature of physical processes like the extent of convection within tropical clouds - a process which drives the transport of heat around the world.

Lowest rise

So no two simulations will produce exactly the same results; overall, the project produces a picture of the possible range of outcomes given the present state of scientific knowledge. [Now the cat is out of the bag ... the truth is known.  This isn't a study that says the temps will rise 11c, but a series of individual studies in which one predicted a 11c increase in temps.]

The lowest rise which climateprediction.net finds possible is 2C, ranging up to 11C.

The timescale would depend on how quickly the doubling of CO2 was reached, but large rises would be on a scale of a century at least from now.

"I think these results suggest that our need to do something about climate change is perhaps even more urgent," the climateprediction.net chief scientist David Stainforth told BBC News.

"However, with our current state of knowledge, we can't yet define a safe level in the atmosphere."

On Monday, the International Climate Change Taskforce, co-chaired by the British MP Stephen Byers, claimed it had shown that a carbon dioxide concentration of over 400 ppm (parts per million) would be 'dangerous'.

The current concentration is around 378 ppm, rising at roughly 2ppm per year.

Dangerous warming

Next week the UK Meteorological Office hosts an international conference, Stabilisation 2005, announced by Tony Blair late last year.

Its aim is to discuss what the term "dangerous" global warming really means, and to look at ways to stabilise greenhouse gas levels.

Myles Allen, the principal investigator of climateprediction.net, said the focus on stabilisation might not be appropriate.

"Stabilisation as an exclusive target may not be adequate," he told BBC News.

"Stephen Byers claims to know that 400 ppm is the maximum 'safe' level; what we show is that it may be impossible to pin down a safe level, and therefore we should not focus exclusively on stabilisation."

Distributed computing has been used before, notably by the Search for Extra-Terrestrial Intelligence or Seti, where several million people have downloaded software enabling them to analyse data from observations of distant galaxies for signs of alien life.

The scientists behind climateprediction.net believe their project, because it is distributed to individual PCs, can help inform people about climate change - and that, in turn could bring political change. [Are they scientist or politicians?  Neither I'd suspect, but a modern day version of John Cumming.]

"It's very difficult to get politicians to collaborate, not only across the globe but also over sustained lengths of time," Bob Spicer from the Earth Sciences Department at the Open University, told BBC News.

"The people who can hold politicians to account are the public; and with this project we are bringing cutting-edge science to the stakeholders, the public."

Computers producing slightly different results does not imply either computer is wrong or malfunctioning.

Let me relate a problem I took on when I was doing my BE Chem Eng , in an elective call Advanced Computations. I was the goal for this project based subject asked to device a program (in Fortran 77) to solve a range of partial differential equations , and then to use the results produced to parametrise the data (get back to the partial differential equations from the data numerically) , I used a mathematical library called the NAG Library to solve the partial differential equations (which weren't in standard forms with known analytic solutions), and then used LQM diagonalisation to parametrise the data - get the coefficients for each term.

I had access to the mathematics and engineering department mainframes and ran my code (identical on both systems) and was surprised to get different answers, disturbing, I had not realised that higher order partial differential equations tend to produce solutions that are complex numbers (and so come in pairs , ie the poles and zeros are complex and include either an +i or -i ( as complex conjugates of each other).
Neither answer was wrong , they were both perfectly acceptable solutions to the differential equation. The same applies ,though the solution is less complicated for ordinary differential equations of order 2 or higher. One or may simply be unphysical but still correct.

So - producing different answers in a massively parallel computing system (over the net or a network) in no way implies either answer produced is incorrect, the answer will actually be a combination of these answers. Anyone with a knowledge of vector calculus and differential equations and boundary value problems , and of physics will know this.

You understanding of the engineering and physics is a deficient.

Your other "objections" are red herrings. :loling:

On the matter of massive parallelism - the following I extracted from my course notes for Elec4700 (2004) which I completed a couple of months ago and is relevent:
Quote:
 
Global Arrays
A vital consideration in using (Massively Parallel Processors) MPPs is how data are stored. So-called replicated-data schemes require that a copy of each data item in the program be stored on each processor, so that the size of the problem that can be handled is limited by the memory of a single processor. In distributed-data applications each processor holds only a part of the total data; in such cases, the problem size is limited only by the total memory of the machine, allowing much larger problems to be treated.

No emerging standards for parallel programming languages (notably just High Performance Fortran (HPF-1)) provide extensive support for multiple instruction multiple data (MIMD) programming .The only truly portable MIMD programming model is message passing, for which a standard interface is now established .It is, however, very difficult to develop applications with fully distributed data structures using the message-passing model .The shared-memory programming model offers increased flexibility and programming ease but is less portable and provides less control over the inter-processor transfer cost. What is needed is support for one-sided asynchronous access to data structures (here limited to one- and two-dimensional arrays) in the spirit of shared memory. With some effort this can be done portably ; in return for this investment, a much easier programming environment is achieved, speeding code development and improving extensibility and maintainability. A significant performance enhancement also results from increased asynchrony of execution of processes ; with a one-sided communication mechanism, where each process can access what it needs without explicit participation of another process, all processes can operate independently. This approach eliminates unnecessary synchronization and naturally leads to interleaving of computation and communication. Most programs contain multiple algorithms, some of which may naturally be task-parallel (e.g., Fock matrix construction), and others that may be efficiently and compactly expressed as data-parallel operations (e.g., evaluating the trace of a matrix product). Both types of parallelism must be efficiently supported. Consideration of the requirements of the SCF algorithm, the parallel COLUMBUS configuration interaction program , second order many-body perturbation theory and parallel coupled-cluster methods  led to the design and implementation of the Global Array (GA) toolkit  to support one-sided asynchronous access to globally-addressable distributed one- and two-dimensional arrays.

This toolkit provides an efficient and portable ``shared-memory'' programming interface for distributed-memory computers. Each process in a MIMD parallel program can asynchronously access logical blocks of physically distributed matrices, without need for explicit co-operation by other processes. Unlike other shared-memory environments, the GA model exposes the programmer to the non-uniform memory access (NUMA) timing characteristics of the parallel computers and acknowledges that access to remote data is slower than to local data. From the user perspective, a global array can be used as if it were stored in shared memory, except that explicit library calls are required to access it. The information on the actual data distribution can be obtained and exploited whenever data locality is important. Each process is assumed to have fast access to some ``local'' portion of each distributed matrix, and slower access to the remaining ``remote'' portion. Remote data can be accessed through operations like ``get'', ``put'' or ``accumulate'' (floating point sum-reduction) that involve copying the globally accessible data to/from process-private buffer space. The toolkit provides operations for (i) the creation, and destruction of distributed arrays, (ii) the synchronization of all processes, (iii) inquiries about arrays and their distribution, and (iv) primitive operations, such as get, put, accumulate, atomic read and increment, gather and scatter, and direct access to the local portion of an array, A number of BLAS-like data-parallel operations have been developed on top of these primitives.

Additional functionality is provided through a variety of third party libraries made available by using the GA primitives to perform the necessary data rearrangement. These include standard and generalized real symmetric eigensolvers (PeIGS, see below), and linear equation solvers (SCALAPACK) . The O(N2)cost of data rearrangement is observed to be negligible in comparison to that of O(N3) linear-algebra operations. These libraries may internally use any form of parallelism appropriate to the host computer system, such as co-operative message passing or shared memory.
Offline | Profile | Quote | ^
 
somerled
Member Avatar
Admiral MacDonald RN
captain_proton_au
Jan 28 2005, 09:23 AM
somerled
Jan 28 2005, 02:45 AM

This is a very good example of the application of massive parallelism , ie lots of CPUs all working at once , via TCP , on the tiny parts of a huge calculation problem , a poor man's SUPERCOMPUTER.


Well thats pretty much how the supercomputers nowadays are constructed, just 5000 cpu's patched together

This is true , and this is why the subject of Massively Parallel Processor Design and Programming was part Elec4700 , it's a natural progression from Pipelined Processor Design which is where the subject spent most of it's time.
Offline | Profile | Quote | ^
 
gvok
Unregistered

Somerled, do you know how far away we are from an "atomic computer" (i.e., one that uses the quantum states of atoms to store memory). BTW I might be a little off with my terminology.
| Quote | ^
 
somerled
Member Avatar
Admiral MacDonald RN
gvok
Jan 28 2005, 10:38 AM
Somerled, do you know how far away we are from an "atomic computer" (i.e., one that uses the quantum states of atoms to store memory).  BTW I might be a little off with my terminology.

It was discussed at the end of the subject , a fair way off before a "generic atomic computer" sits on any of our desks :( . Most of the research is focusing on quantum computing and getting quantum NAND gates and LATCHES working.

The beauty of quantum computing is that you can simultaneously do two things at once with the one gate (or get double the results from a latch). :wow:



NAND gates and LATCHES are the basis for all computing.



Maybe our grandkids will have them.
Offline | Profile | Quote | ^
 
gvok
Unregistered

Iteresting. Thanks.
| Quote | ^
 
ds9074
Member Avatar
Admiral
They had one of the researchers on the BBC. He was saying that while the media has picked up on the headline 11 degree figure, which is the top end of end of projections, the more interesting figure for him was the minimum figure. In no simulation was there warming of less than around 2 degrees. That is still enough to cause substantial climate change.
Offline | Profile | Quote | ^
 
Admiralbill_gomec
UberAdmiral
The two degree figure is rarely mention for the reason that it could be explained away by a margin of error. So, you make the figure as large as possible and if nothing happens you simply say, "Well, it was just a model. It just shows what COULD happen."

Food for thought.
Offline | Profile | Quote | ^
 
ds9074
Member Avatar
Admiral
The researcher was actually saying that the 2 degree minimum figure was the most interesting because it showed that in all projection the earth is going to warm, we are going to see global warming, its just a matter of how much not if.
Offline | Profile | Quote | ^
 
Dwayne
Profanity deleted by Hoss
somerled
Jan 28 2005, 10:27 AM
Computers producing slightly different results does not imply either computer is wrong or malfunctioning.

Can anyone point out where I suggested or stated the computers were wrong or had malfunctioned?

somerled
Jan 28 2005, 10:27 AM
You understanding of the engineering and physics is a deficient.

Your understanding and comprehesion of the English language is what is deficient, and your inane ramblings about "partial differential equations" really highlights your egotistical and narcissistic behavoir.

somerled
Jan 28 2005, 10:27 AM
Your other "objections"  are red herrings.

The real red herring was your entire reply to my post.
Offline | Profile | Quote | ^
 
somerled
Member Avatar
Admiral MacDonald RN
Dwayne
Jan 28 2005, 07:55 PM
somerled
Jan 28 2005, 10:27 AM
Computers producing slightly different results does not imply either computer is wrong or malfunctioning.

Can anyone point out where I suggested or stated the computers were wrong or had malfunctioned?

somerled
Jan 28 2005, 10:27 AM
You understanding of the engineering and physics is a deficient.

Your understanding and comprehesion of the English language is what is deficient, and your inane ramblings about "partial differential equations" really highlights your egotistical and narcissistic behavoir.

somerled
Jan 28 2005, 10:27 AM
Your other "objections"  are red herrings.

The real red herring was your entire reply to my post.

:loling:

Oh . So you didn't follow the brief discussion on solving and parametrising PDEs and BVPs ? That only indicates your knowledge of calculus and physics is woefully inadequate , and you really don't have a clue.


:rotfl:

More red herrings from the dwayne that are totally irrelevent to the topic in hand. Now isn't that typical ?



Others :

You might find the visualisations Visualizations of simulation results - Climate Projections by Deutsches Klimarechenzentrum GmbH interesting , in particular the temperature change due to greenhouse gas emissions with a coupled atmosphere ocean model . (Note it's a 17Mb MOV.file so it may take a while to download if using dialup.)

Here are more links that might be of interest. I'm sure there is lots of pertanent information to be gleened - no - I haven't looked at every link given.Climate Links

And wrt to uncertainties in the model , here are some plots
Plots of uncertainty on projected temperature rise
of most interest are
Uncertainty in predictions of the climate response to rising levels of greenhouse gases
The Simulations.
Anomality fields - global
Posted Image
which is the distribution of modeling projected temperature increases. The most likely change in temperature predicted is around 4 degrees C, with an uncertainly somewhat less than +/- 0.8 degrees C at the peak value(see above)
Offline | Profile | Quote | ^
 
Dwayne
Profanity deleted by Hoss
somerled
Jan 29 2005, 12:13 AM
Oh . So you didn't follow the brief discussion on solving and parametrising PDEs and BVPs ? That only indicates your knowledge of calculus and physics is woefully inadequate , and you really don't have a clue.

Another somberlad red herring and more proof you're egotistical.

It's so sad you have to play these little pathetic games.
Offline | Profile | Quote | ^
 
somerled
Member Avatar
Admiral MacDonald RN
Dwayne
Jan 29 2005, 12:34 AM
somerled
Jan 29 2005, 12:13 AM
Oh . So you didn't follow the brief discussion on solving and parametrising PDEs and BVPs ? That only indicates your knowledge of calculus and physics is woefully inadequate , and you really don't have a clue.

Another somberlad red herring and more proof you're egotistical.

It's so sad you have to play these little pathetic games.

Some of us are actually interested in discussing the topic , so how about either saying something relevant or refraining from trying to bait people .

Wicheta - are you going to council Dwayne on this or are you going to permit his nonconstructive posting to persist ? Many of us are getting pretty bored and pissed off with his tactics.
Offline | Profile | Quote | ^
 
ZetaBoards - Free Forum Hosting
Free Forums with no limits on posts or members.
Learn More · Register Now
Go to Next Page
« Previous Topic · Politics and World Events Forum · Next Topic »
Add Reply

Tweet
comments powered by Disqus