Slashdot Mirror

Fine, next time antarctica researchers need medical equipment and rescue missions from the government, screw 'em. It's not the rest of the world's job to be their backup plan.

i found some science info

http://lamar.colostate.edu/~hillger/flashcardsbasi c.pdf

http://www.pnl.gov/ag/usage/acroel.html#k

there is no such shit as "KPH"

See http://www.netlib.org/benchmark/performance.pdf page 53.

1. Earth simulator
2. ASCI Q
3. Virginia Tech G5 cluster (9.555 Tflops and rising, $5.2M HARDWARE ONLY)
4. PNL Itanium2 cluster (8.633 Tflops, $24.5M HARDWARE ONLY)

So nope, not only will the PNL Itanium2 cluster not be #2, it will also be 1Tflop behind the Virginia Tech cluster, and it will have done it at almost 5 times the cost. Bravo!

And this is a photograph of "the 2 metric ton accumulation of melted nuclear fuel dubbed the "Elephant's Foot" located in Corridor 217/2.". HAND.

ASCI White (#4) is at LLNL, but it is not a linux cluster- perhaps that contributed to the confusion.

The largest Linux supercomputer group in the NorthWest, and in the World. Which is located Less than 5 hours away from Bellingham, in southeastern Wasington at PNNL. I have not heard of any invitation to present at Linuxfest. Seems like this is a great thing to be shown/talked about at LinuxFest Northwest....

The image shows some guy calibrating the detector. The large black cylindrical part is a very heavy plastic detector.

Notice how the Pythagorean Theorem...

Like this?.

Took about 30 seconds with google, and that's because I misspelled Pythagorean. Good thread, however.

Check AWU about the possibilities at these facilities.

Also, check these:

Sandia

Los Alamos

Argonne

Brookhaven

Pacific Northwest

Lawrence Berkeley

Lawrence Livermore

Oak Ridge

And there are other other national labs that I did not mention.

• a network of point-to-point "condominium" wavelengths
• condo owners can recursively partition their wavelengths
• wavelength owners determine topology and routing of their light paths
• massive edge peering, "star bursts" vs. "ring of rings"
• not "distributed network objects", but "distributed object networks"

• customer owns infrastructure, carrier provides network management
• asset-based telecom allows customers to fund and control the network
• customer controls the bandwidth

• DoE presentation (2mb pdf)
• 3rd-gen optical network, CANet*3
• Further reading

I am currently involved with the development of a middleware system called Scientific Annotation Middleware - SAM. One of the services that we are in the process of implementing is a Notorization Service that can be used by a 3rd party for signing document hashes. We use the XML signiture spec./infrastructure.

In particular we'll be developing Notebook services and a SAM electronic-notebook that will use the notorization service for exactly the purpose you seek.

Unfortunately, it won't be viable/released for end-user use for at least a couple more years.

There are other e-notebooks that have been developed (by us and other parties), but none of them have legally acceptable notorization capabilities to date.

What did you guys do with all those harding working Pigeons?!

Seriously though... can they actually do mork work than than these penguins?

We have a group dedicated to cybersecurity.

That was my first thought. We have to redefine the term supercomputer. It used to be that these massively parallel systems were connected through high-bandwidth memory.

According to the detailed article, it is made up of 700 nodes connected to a SAN. Hence, the term supercomputer seems to cover a network of computers.

Given that broader definition, one could argue that metaprocessors like United Devices, SETI@Home, distributed.net are supercomputers. Although their connectivity is poor, they have a lot more than 1400 processors.

I had the fortunate experience of attending a supercomputing conference in 1995 while working for HP. This was back when Thinking Machines, etc. were in their heyday. At the end of the conference, we talked about future supercomputers. Everyone conceded that the internet was the supercomputer of the future, where people would lease out compute power.

I think what we are seeing now is the nascent stages of this technology. Idealistic research projects may give way to capitalistic projects. Imagine getting paid to share your computer. Take the $24.5M and pay people $50 a system. You could get 490,000 systems at your disposal. :-)

http://www.emsl.pnl.gov:2080/docs/collab/
i worked on this as undergrad, its been along going steadily for years. There are other DOE notebook projects at Berkeley and Oak Ridge--they all supposedly can share data with each other.
...in the post preview section slashdot seems to be peeling off the :2080 in my URL for some unknown reason. If you can't get to the site paste the link in and it should work.

There has been lots of work on information visualization. It's not really like The Matrix, but attempts to be. Check out PNL's visualization team's work. In Citeseer, look for articles about Bead and Lyberworld.

I've done some work on this with a system called Yavi; can send a reference if you want. It's not hard to find lots of work on this. For an historical bibliography (1993 + prior), see my references here.

In a typical group of 100 signal lights with 120-watt lamps, the electrical load is 12,000 watts (12 kilowatt, kW). If the operating hours are 6,132 hours per year, the electrical consumption will be 73,584 kilowatt-hour (kWh), resulting in an annual electrical cost of $3,776. The use of LED technology reduces this cost by 85%. The lamp rating of the LED lamp is 17 watts, resulting in an electrical load of 1,700 watts (1.7 kW). Using the same operating hours, the electrical consumption will be 10,425 kWh, resulting in an annual electrical cost of $534 (thus the 85% savings).

The normal lamp-life of a 120-watt lamp is 8,000 hours. For the traffic signals described above, the lamps would have to be replaced every 1.3 years. At this time, the replacement cost and the maintenance cost are realized. The lamp life for LED technology is 100,000 hours. Therefore, the replacement period extends to 16.3 years, resulting in a significant cost savings for lamps and maintenance cost.

Check it:
Fort Hood Converts to LED's

While Mr. Lanier's article was both interesting and insightful (as all those moderators say), I cannot agree with his statement about Artificial Intelligence.

Mr. Lanier argues that because humans have not in the past and cannot currently develop computer programs that duplicate the brain processes, that artificial intelligence cannot be achieved. He fails to take into account various research being performed with neural net and related architecture, and further development into our understanding of the human brain. While I agree with most of his arguements, his reasoning behind artificial intelligence is not one of them.

J1

Having once worked at a National Lab (PNNL, operated by Battelle) I can say that if LANL is anything like where I worked, they're not lying. Nuclear material is handled very carefully and a forest fire is the least of their worries.

There are other issues with the Natl labs (namely, national security) that aren't all that great - but I don't think fire is going to be a problem.

(yes, I was certified as a level 1 rad worker, and no it wasn't my main focus - I was a computer tech who had to go into rad zones from time to time. if you think an old Quadra 605 is slow now, it feels ten times as slow when you're surrounded by geiger counters and have a quota as to how long you would be allowed to stay there)


- Jeff A. Campbell
- VelociNews (http://www.velocinews.com)

I think I'd rather help bring it around than just sit and hope.

An excellent area for contribution is design software. Currently there are a number of excellent free molecular modeling packages: MMTK, NAMD, Moldy, NWchem. There are also several excellent display programs: RasMol, VMD, Midas, and my own feeble effort, xyz2rgb. What is still lacking is:

• Software to generate structures painlessly. Two efforts in this area are CavityStuffer by Markus Krummenacker, DiamondCAD by Chris Phoenix and John Michelsen, and some tinkering of mine.
• Some kind of wrapper that makes all this stuff easy to use. There is a commercial package called HyperChem, and the DiamondCAD folks are working on an open-source version called OpenChem.

• University of Toronto Neural Network Group has some software which fits the Open Source model.
  
• University of Stuttgart has a neural network simulator. Source is available, its GPL.
  
• Pacific Northwest National Laboratory has a slew of links to free/share ware. No idea if any of it is Open Source.
  
• Karsten Kutza maintains Neural Networks at Your Fingertips.