• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans


Adverts help to support the work of this non-profit organisation. To go ad-free join as a Member.


Photo
- - - - -

Global Grid


  • Please log in to reply
26 replies to this topic

#1 manofsan

  • Guest
  • 1,223 posts
  • 56

Posted 17 November 2004 - 01:50 PM


IBM has set up the Global Community Grid (gee, that title sounds like a ripoff of WorldWideWeb)

http://www.worldcomm...w/join_now.html

The Grid will use idle computer time to solve protein folding problems and other useful stuff.
Sounds pretty nifty, so join up if you like.

#2 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 17 November 2004 - 03:43 PM

Interesting, sounds similar to Folding@home.

http://folding.stanford.edu/

Folding@Home is a distributed computing project which studies protein folding, misfolding, aggregation, and related diseases.

Reason has a concise description and instructions about joining the Longevity Meme Folding@home team here:

http://www.longevity...ing_at_home.cfm

IBM recently announced (Nov 5) their Blue Gene supercomputer computer broke the worlds record held by Japan's NEC Earth Simulator. The NEC supercomputer clocks up to 35.86 teraflops (trillion floating point operations per second). Blue Gene came in at 70.72 teraflops.

http://news.bbc.co.u...ogy/3983131.stm

Folding@home is probably the worlds most powerful computing system, currently running at ~193 TFlops

http://vspx27.stanfo...y?qtype=osstats

sponsored ad

  • Advert

#3 ocsrazor

  • Guest OcsRazor
  • 461 posts
  • 2
  • Location:Atlanta, GA

Posted 17 November 2004 - 04:29 PM

One of my long-term projects is to devise a project similar to these to do neuro-modeling of large cortical networks and neural signal processing. We already have supercomputer time available here at Georgia Tech, but I can forsee that we may eventually need a much larger amount of processing power. I thought of writing up a proposal for Google Compute, much like Folding@home has done.

#4

  • Lurker
  • 0

Posted 18 November 2004 - 12:21 AM

I love the idea of distributed computing.

I've envisioned an idea by which the bittorrent file sharing system, arguably one of the most prominant network of people sharing files in the world, could trade a percentage of their upload bandwidth (per second) for a percentage of their computing power which would go to one or a number of distributed computing programs. It's just an idea but it gives direct incentive for those who do not want to share their entire upload bandwith with others for various reasons. I'm not sure if it's directly compatible with bittorent but the idea seems workable and applicable to some sort of filesharing system. The key being there is direct incentive in offering up some computing power.

It's much cheaper and easier than having to build a supercomputer yourself. The users of your distributed computing programs upgrade their computers themselves at their discretion, at no cost to you.

#5 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 18 November 2004 - 02:23 AM

Yes, manofsan, and cosmos,

With distributed computing, previously intractible problems become feasable. It's an excellent example of pushing the envelope of progress.

Looks like SETI@home ranks up there.

SETI@home is a scientific experiment that uses Internet-connected computers in the Search for Extraterrestrial Intelligence (SETI).

http://setiathome.ssl.berkeley.edu/

SETI@home is currently running at ~ 66 TFLOPS

http://setiathome.ss...edu/totals.html

-----------------------------------------------------------

My computer is folding proteins with the Longevity Meme team.

http://www.longevity...ing_at_home.cfm

#6

  • Lurker
  • 0

Posted 18 November 2004 - 04:01 AM

Computing power and application are among the most tangible measures of human progress.

dnamechanic, that 193 teraflop quote makes that distributed computing program have more computing power available to it than any supercomputer built yet.

#7 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 18 November 2004 - 10:49 PM

Yes, cosmos

The scientists at Stanford, and all the individual contributors, have developed a very valuable system (Folding@home).

IBM is working on a faster version of the Blue Gene supercomputer.

"Blue Gene L, which will deliver between 180 teraflops and 360 teraflops, will cost between $50 million and $100 million to complete, or about $200,000 per teraflop"

http://www.pcworld.c...d,113418,00.asp

Accordingly, Folding@ home is providing a value worth more than $100 million.

Folding@home's ~193 teraflops of computing power is the output of ~170,000 CPU's.

There are about 660-670 million PC users today according to CNET news (August 2, 2004). "More than a 152 million PCs will leave factories this year." "By the end of the decade, a billion people will be clicking away at computers."

Most of those computers will have excess capacity.

Obviously, IBM can see this potential resource in the Global Community Grid.

The world computing capability, from personal computer CPU's, is growing at a rate equivalent to approximately 1,000 Folding@home supercomputers per year.

This capability, if utilized, can help us reach our goals.

#8 123456

  • Guest
  • 295 posts
  • 0

Posted 19 November 2004 - 01:44 AM

I just signed up with "http://www.worldcommunitygrid.org/join_now/join_now.html".
I will try to do what I can help reach the goal of immortality. Thank you for mentioning this site to us manofsan.

Edited by 123456, 19 November 2004 - 04:26 AM.


#9 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 23 November 2004 - 01:32 AM

Good move, 123456

Your computer is working on "The Solution".

#10 manofsan

  • Topic Starter
  • Guest
  • 1,223 posts
  • 56

Posted 23 November 2004 - 01:58 PM

If someone could develop an open-source platform for this distributed computing grid, then the field would be open for anyone to develop grid computing applications. Gee, I wonder if struggling companies like Sun would be interested in pursuing this, since they're the ones who developed Java, etc.

In order to avoid fraud and misuse by hackers, etc, the open-source platform would include domain-key certification, so that applications could not be anonymously distributed. So perhaps it could be a suped-up version of Java, but specifically specialized or streamlined for math computation.

What do you all think?

#11 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 27 November 2004 - 06:10 PM

Yes, manofsan,

Sun is an innovative company.
Sun's Java is an example of a successful product that has yielded little profit.

An open-source platform would certainly speed up grid computer utilization.
Corporations and institutions with large numbers of mostly idle PC's may be interested in such a product.

Hopefully Sun is interested. If not some other group else will probably develop it.

The worldwide prevalence of mostly idle computers is a significant untapped resource.

--------------------------------------------------------------------

To achieve our goals, we need all computer processing power available.

---------------------------------------------------------------------

One of the most computational intensive problems in biology is the modeling of proteins.

Proteins are the nano-machines that carry out instructions coded in our DNA.

Understanding the structure and function of proteins is a key to reaching our goal.

"Examining the entire human genome could require up to 1,000,000 years of computational time on an up-to-date PC. Using a commercial 1000 node cluster would require 50 years and, while faster, would still be impractical."

"Only the grid will allow the scaling of this total procedure... to genomes
as large as the Human Genome."

http://www.grid.org/...s/hpf/about.htm

Hopefully, institutions like Stanford University (Folding@home) and forward-looking
corporations like IBM (Global Community Grid), have shown the way to solving this massive problem.

Right now, everyone with a computer has the capability to contribute to this key problem in understanding disease and life itself.

A contributor can join a team of like-minded contributors.
This helps to keep the effort in perspective.

The Extropy Institute is an example, many contributors are also members of ImmInst.

http://vspx27.stanfo...age&teamnum=346

There are many teams. One can start a new team.

Mentioned in a previous post:

Reason, the founder of The Longevity Meme has concise description here:

http://www.longevity...ing_at_home.cfm

------------------------------------------------------------------

Enjoy folding with the Grid, or F@H, or with others.

It must be done, sooner is better.

#12

  • Lurker
  • 0

Posted 27 November 2004 - 07:10 PM

There are a few key achievements that would contribute to the worldwide mainstream recognition and use of distributed computing programs.

Among them (in my opinion):
- using file-sharing networks to create an incentive driven push to sharing your computer power (I described this in my other post)
- modest advertising to niche markets of people with fast computers, conducive mindset, budget for advertising could be diverted from possible future supercomputing projects
- other incentive driven endevours, for instance computing teams that achieve certain goals are recognized and perhaps offered some sort of collective prize

Collective computing power can be shared among multiple projects including:
- human genome analysis
- astrophysical computational work (everything within this broad area)
- SETI@home
- various drug research, disease research projects

#13 lightowl

  • Guest, F@H
  • 767 posts
  • 5
  • Location:Copenhagen, Denmark

Posted 27 November 2004 - 08:38 PM

A couple of friends and I have started a team called "AI". You are all welcome to join us.
EDIT: http://www.worldcomm...mId=D1DS1NR1BN1

I like the competition element of these grid systems. People are actually competing to do good. ;)

It will be interesting to see what new projects other than molecule folding IBM is going to consider. They are saying they will start 6 new projects per year.

Any ideas for a project from ImmInst?

I had the idea to help the researchers at Methuselah Mouse Prize with information, but I am no biologist. Does any of you biology guys know how this could be done?

Edited by lightowl, 28 November 2004 - 06:16 AM.


#14 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 27 November 2004 - 09:42 PM

Good comments, cosmos.

Anything that helps the collective computing cause is welcome.

1. The file-sharing idea is a good one. An organization with a large upload capability and a product to offer clients, could trade for excess CPU capability of clients.

2. Advertising is good idea, especially for corporations with the organization and budgets for large supercomputing projects.

Advertising is a form of education and that would greatly help.

Potential contributors would be interested to know that grid computing:

- is typically set up such that it uses CPU capacity that is not being used, i.e. doesn’t slow the computer down.

- requires uploading/downloading only when receiving a new part of the task or sending completed results to the central server.

3. The idea of a prize or recognition seems appealing.
Prizes give a huge return for investment.

“The $10 million Ansari X Prize for suborbital flight attracted more than $160 million in research funding and grew an industry in seven years.”

From: http://www.methuselahfoundation.org

Earlier this year, DARPA’s Grand Challenge offered $1 million prize. This prize attracted over 100 teams. One team spent more than $3 million competing.

http://www.spectrum....04/0604dar.html

-------------------------------------------------------------------------

Professor Vijay Pande, the principal investigator in charge of Folding@home, gave a presentation entitled:

Folding@Home: Can a grid of 100,000 CPUs tackle fundamental barriers in molecular simulation?

The video of this seminar is archived at:

http://www.parc.xero...icle.php?id=274

In response to a question from the audience, he said the organization could not afford to pay contributors for use of their computers.

His comment is quite logical, because the value is high.

We are contributing something of significant value.

#15 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 27 November 2004 - 09:43 PM

Hey lightowl,

AI team, as in Artificial Intelligence? Good choice.
I’m surprised the name wasn’t already taken.

Yes, interesting, the new projects from IBM will reveal what IBM perceives as top priorities.

The protein-modeling problem is a vast one, and very worthy of several supercomputer efforts.

Understanding the nano-machines (proteins) of biology is crucial to efforts to greatly extend our healthy lifespan.

You have good questions.

#16 manofsan

  • Topic Starter
  • Guest
  • 1,223 posts
  • 56

Posted 29 November 2004 - 01:29 PM

IBM, Sony, Toshiba have announced the Cell processor will debut early this coming year:

http://www.physorg.com/news2147.html

This is their much-awaited processor which will be stackable for computing and also embedded into consumer electronics, including the Sony Playstation3.

It is a multicore 64-bit processor, designed for heavy floating point and high-bandwidth communication in order to facilitate multiprocessing stackability. Just what the doctor ordered.

Now this kind of processor would greatly help distributed computing. It'll be in consumer electronics as well, so even your TV set could help out F@H when it's idle.

#17 manofsan

  • Topic Starter
  • Guest
  • 1,223 posts
  • 56

Posted 29 November 2004 - 03:33 PM

More on the Cell processor [Heh, what a fitting name ;) ]

http://www.eet.com/s...icleId=54200580

As you can see, Cell is made for multiprocessor networking. I bet Google might like this stuff. They use words like "apulet" instead of applet, but you get the idea -- it uses total encapsulation to ensure process integrity.

This architecture has high scaleability, however as commented on at the end of the article, it has memory bandwidth limitations as the bottleneck.

#18 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 04 December 2004 - 11:40 PM

Yes manofsan,

That is good info on the "Cell Processor".

You are probably right about Google being interested.

The Google team is the top contributing team to Folding@home.

http://vspx27.stanfo...qtype=teamstats

Google promotes this research effort via the Google toolbar.

http://toolbar.googl...dc/offerdc.html

In the future Google may work with other research teams on similarly complex problems.

http://toolbar.googl..._dc.html#about8

For those interested in grid computing and how P2P applications such as KaZaA fits into the picture, there is a good article entitled:

Perspectives on grid: Grid computing -- next-generation distributed computing

http://www-106.ibm.c...ry/gr-heritage/

Grid computing applications include:

- Processing seismic data in the search of petroleum
- Financial services at Charles Schwab & Co.
- Breast cancer screening and diagnosis (University of Pennsylvania)
- and many others

http://www-1.ibm.com...s_stories.shtml

Seems as though the Cell processor or something similar will help us solve some of the currently intractable problems.

In the meantime, the nano-machines (proteins) of biology have much to teach us.

Proteins are assembled, mostly linearly, one amino acid at a time by the instructions coded in our DNA.
These strings of amino acids then fold into a three dimensional conformation, where they take on their role as a nano-machine.

This folding process is exceedingly complex. But, to fully understand the relationship between DNA and a functional protein (biological nano-machine), one must understand the folding process.

When the nano-machines of biology are better understood, we have a chance of emulating or improving upon them via emerging nanotechnology.

So let's keep our computers Folding.

Edited by dnamechanic, 05 December 2004 - 09:24 AM.


#19 manofsan

  • Topic Starter
  • Guest
  • 1,223 posts
  • 56

Posted 08 December 2004 - 10:01 PM

Big players getting together to do Grid:

http://www.ecommerce...Grid-38806.html

Geez, I think this is all going to get bogged down in proprietary standards, like a big Tower of Babel.

Oh well, if Grid takes off, it'll take a lot of the pressure off the standards wars, by providing a common format for interoperability.

#20 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 11 December 2004 - 09:30 AM

Interesting article on the Megagrid, manofsan.

As the article states "Project MegaGrid has a mega-goal". They are trying to develop a standard approach to building and deploying an enterprise grid computing infrastructure. Nice to see the big name corporations.

The guys at usenet newsgroup; amd.alt.comp.hardware.overclocking,

informed me that the University of Oxford has teamed with the Natural Environmental Research council (NERC), and other prestigious organizations to create climateprediction.net.

http://www.climateprediction.net/

Their objective is to try to forecast the climate in the 21st century.

As most of you know, climate prediction is complex and climate prediction estimates have only been the in the domain of supercomputers.

Most scientists believe the earth is likely to warm in the near future, the question is how much? According to climateprediction.net, "Using the internet and computers of the general public is the only possible way of running all the models necessary to gain an understanding of this vast problem".

The climateprediction.net project is using the BOINC (Berkeley Open Infrastructure for Network Computing) software platform, developed by scientists at the University of California in Berkeley.

"It’s the perfect application for public distributed computing” said Dr. David Anderson, Director of the SETI@home and BOINC projects."

This is another project important for our future.

We have unused computer power available now. I say "full speed ahead", to those that can mobilize this huge resource.

Seems like the computer hobbyists have an excellent perspective on these important issues. If one looks at the major contributors to these grid projects one will see computer hobbyists as significant contributors.

Take a look at the top contributors to Folding@home:

http://vspx27.stanfo...qtype=teamstats

To a large degree they consist of such groups as: Overclockers Australia, [H]ardOCP, ...and how about this one "kcgeek".

These gamers and computer hobbyists are helping us in two ways: They help drive computer development, and by participating in grid projects, they are also a significant factor in helping us to reach our goals.

#21 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 18 January 2005 - 03:26 PM

The Folding@Home computer system at Stanford University significantly expanded their projects this past year.

Results from a Folding@Home cancer project were recently published in the Journal of Molecular Biology.

http://www.ncbi.nlm....t_uids=15588832

F@H website: To our knowledge, this is the first peer-reviewed results from a distributed computing project related to cancer....Roughly half of all known cancers result from mutations in p53. Our first work in the cancer area examines the tetramerization domain of p53. We predict how p53 folds and in doing so, we can predict which amino acid mutations would be relevant. When compared with experiments, our predictions have appeared to agree with experiment and give a new interpretation to existing data.

--------------------------------------------------------------------------------------
F@H also recently broke the 200 TFLOP number, currently at 201.8 TFLOP with over 175,000 active processors.

#22 123456

  • Guest
  • 295 posts
  • 0

Posted 06 June 2005 - 01:23 AM

World Community Grid

Currently have 8 members representing Imminst.

Did you guys stop participating?; No one else crunching recently it seems.

#23 Lazarus Long

  • Life Member, Guardian
  • 8,116 posts
  • 242
  • Location:Northern, Western Hemisphere of Earth, Usually of late, New York

Posted 06 June 2005 - 01:27 AM

I was a folding at home volunteer until the conflicts between firewalls, them, and my kids trying to access places they shouldn't go crashed and burned my computer.

I ran into trouble with also not being able to put my machine in standby so after I rebuilt the hard drive I did not rejoin. After I get a few machines going at the same time I might designate one to be online 24/7 and replace that one with them.

#24 eternus

  • Guest
  • 6 posts
  • 0

Posted 12 June 2005 - 12:25 AM

http://www.worldcomm...w/join_now.html
My computer now folds proteins in the name of the greater good.

#25 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 12 June 2005 - 01:25 AM

The Longevity Meme has an active Folding@Home team with increasing numbers of participants.

Reason reports here:

http://www.longevity...fm?news_id=1695

The progress of the Longevity Meme is being tracked at two sites.

Statsman shows competing teams within plus or minus four places:

http://www.statsman....tats/32461.html

Extreme Overclocking site shows threats, overtakes, and much more:

http://folding.extre....php?s=&t=32461

These groups promote competition and create a massive game, thus encouraging the advancement of science.

There is much friendly competition between teams.

For example a system of computers called the ECU-SCIS-Cluster (actually about 30 computers clustered at a university in Australia) passed The Longevity Meme (TLM) team a while back. This cluster was referred to as "The Borg", primarily because it is an automated computing cluster.

The TLM team raised more participants and has since passed "The Borg".

Tho maybe not for long, "The Borg" has more computers that could be added to the cluster.

We could use some more processing power ;)

http://www.longevity...ing_at_home.cfm

#26 manofsan

  • Topic Starter
  • Guest
  • 1,223 posts
  • 56

Posted 19 June 2005 - 08:48 PM

Here's another article for Grid, but it's about using game consoles:

http://www.wired.com...6,67835,00.html

This is clever because consoles are going to be unused for most of the day, plus there's a huge number of people who buy consoles, and the fact that the new models will have major processing power.

sponsored ad

  • Advert

#27 dnamechanic

  • Life Member
  • 1,518 posts
  • 0
  • Location:Dallas, TX

Posted 19 September 2005 - 01:51 PM

More grid projects:

The Einstein@Home project uses the idle CPU time of your computer to search the
sky for evidence of gravitational waves.

Gravitational waves could result from stars exploding, black holes engulfing each other,
or other acts of astronomical violence.

Article at The Institute of Electrical and Electronics Engineers (IEEE):

http://boldfish.ieee.../u/834/06712426

"The software for Einstein@Home is called BOINC, a whimsical acronym for
Berkeley Open Infrastructure for Network Computing. It's the brainchild of
IEEE Member David Anderson, project director at the Space Sciences Laboratory
at Cal-Berkeley...

...The long-term goal of BOINC is to get everybody in the world directly involved
in volunteer computing for science."

The group has good participation:

"Since going public in mid-February, Einstein@Home now has some 45,000 participants."




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users