Slashdot Mirror

You can't just set up computers in any old warehouse, you need the proper power, air conditioning systems, cable conduits, etc...

Acutually, you can.

Their bottom line to the research: "A well-balanced architecture is nearly insensitive to communications overhead. By contrast a system with weak communications can lose half its power for applications where communications are important."

Conclusion: "For most large scientific and engineering applications the performance is more determined by parallel scalability and less by the speed of individual CPUs. There must be balance between processor, interconnect, and I/ O performance to achieve overall performance. To date, only a few tightly-coupled, parallel computer systems have been able to demonstrate a high level of scalability on a broad set of scientific and engineering applications."

Actually that is not necessarily always true lately.

For instance, these days submissions in Physics happen by people writing a paper, and uploading it to arXiv -- where it gets peer reviewed and you get inputs. And more importantly, you establish in public that you were the first person to come up with FOO.

Not only that, the process is a lot more open than it used to be. Although some J Random dude at Physical Review can (and will) reject your paper for unknown reasons, it's quite unlikely that would happen if it has received excellent reviews after it's put up at arXiv. The process is a whole lot more transparent.

And since only the final editions go to the journals, the paper is still available at arXiv. And arXiv has been working on making several other publications available online - however, this has only begun for papers from and after 1992, so that is indeed a problem.

However, although arXiv does have a CS section, it's not frequented as much as the physics or mathematics sections. Which is a sad thing, IMHO.

And oh btw, in arXiv - the authors own the copyrights, so no question of the journal asking the arXiv to retract the papers. In fact, sometimes authors post their papers after acceptance for a journal publication.

Not too sure about conference publications, though.

You're wrong.
Not just about the correct spelling, but about 'only in America'. In fact, Canada uses that spelling as well. Of course, I am not correcting your spelling of aluminum, as I realize that in other parts of the world, that is the accepted spelling. Wow, funny how that works, isn't it? Realizing that other parts of the world might do things differently? What a fucking concept! Or do you insist that Americans spell curb 'kerb' and tire 'tyre' as well? To-MAY-to or To-MAH-to? Oh fuck, let's call the whole thing off.

I would counter a few of your underlying assumptions with the following references - note that electron tunneling pathways affect protein folding dynamics and that quantum interference plays a critical role in photosynthesis. See also Zeilinger's biomolecule matter-wave interference experiments.

Of course, future computing architectures can incorporate these 'spooky' features.

I would counter a few of your underlying assumptions with the following references - note that electron tunneling pathways affect protein folding dynamics and that quantum interference plays a critical role in photosynthesis. See also Zeilinger's biomolecule matter-wave interference experiments.

Of course, future computing architectures can incorporate these 'spooky' features.

I would counter a few of your underlying assumptions with the following references - note that electron tunneling pathways affect protein folding dynamics and that quantum interference plays a critical role in photosynthesis. See also Zeilinger's biomolecule matter-wave interference experiments.

Of course, future computing architectures can incorporate these 'spooky' features.

Especially the part where it says "The chemical and radiological toxicity of plutonium should be distinguished from each other"

Radium is a common heavy element found in topsoil. Indeed, in high concentrations it is extremely dangerous. Radium is not plutonium, however. I refer you to the Periodic Table. Look for Ra (number 88) and look for Pu (number 94). Notice how they're not the same. In fact the degree of differences between the two elements is comparable to that of Calcium and Iron.

From the theorist's perspectice it doesn't really matter how you implement this stuff - if it works, all implementations are equivalent.
But of course ther are (and will remain) technical advantages of certain implementations. I do not think that currently anybody knows what the most promising physical system is. Trapped ions are probably most advanced at the moment. Compared to them neutral atoms in optical lattices might two advantages: optical lattices appear to be rather "scalable", i.e., one might go beyond 5 qubits rather quickly, once complete coherent control has been demonstrated. (In a linear ion trap there will be difficulties to go beyond 10-20 ions, though very promising ways around these difficulties have also been demonstrated.) On the other hand, using neutral atoms (rather than charged ions) may make the qubits less susceptible to stray fields and other sources of decoherence.

Some early artificial hearts were powered that way as well.

"During the same period, Los Alamos developed a medical-grade fuel for use in cardiac pacemakers and early artificial hearts. The fuel in the artificial hearts was made of 90 percent enriched plutonium-238 and provided up to 50 watts of power. Some of the early pacemakers are still in use; some have been returned to Los Alamos' Plutonium Facility for recovery. " Link

Since alpha rads don't penetrate the skin, it's safe, as long as it doesn't fragment and get carried through the body. Still pretty scary though.

Almost all the links and information contained in the BEAM link from the parent, are broken. I would like to build a Dizzy with a GPS :D

That would be Mark Tilden. He is primarily responsible for the BEAM (current site here) robots, lots of which are based on the bicore circuit and other novel simple circuits(read at the sites!).

Rather than a top-down method of robot design, where you take a brain and make it do cool things, his idea is to decide some basic function you want to accomplish--like walking--and design the simplest circuit that is able to do that. Then pick a new function (say, seeking light) and build that circuit. Hook the two together, and now you have a walking robot that seeks light. It is a more biological method to robot evolution.

If you like the idea of making a functioning insect-robot out of the parts from an old Walkman cassette player and a pocket calculator, you'll love BEAM robotics.

Jim

The only newsy thing about this is the interest from makers of space probes; the thermoacoustic engine has been around for a while (combine with a thermoacoustic chiller and you've got a gas-liquefaction system with no moving parts; here's another one from 1999) and the page from LANL on thermoacoustic systems is almost two years old already. These guys were plugging their sound-to-electricity converter some time ago.

The only newsy thing about this is the interest from makers of space probes; the thermoacoustic engine has been around for a while (combine with a thermoacoustic chiller and you've got a gas-liquefaction system with no moving parts; here's another one from 1999) and the page from LANL on thermoacoustic systems is almost two years old already. These guys were plugging their sound-to-electricity converter some time ago.

You might find this interresting:

http://xxx.lanl.gov/pdf/gr-qc/0104064

-K

The thing that makes Quantum Computers a bit scary is that algorithms like Shor's have shown that QC's can give an exponential speed-up to some problems (assuming P!=NP). Prime factorization happens to be one example that is of particular interest to cryptographers. The reason this is possible is because most of our algorithmic complexity analysis to date is based on the assumption that the computers the algorithms are running on are (read: can be modelled as) Turing machines. Quantum computers are not Turing machines, which is why they get to "break" the rules.

Here's another example of an algorithm for which Quantum computers "break" classical algorithmic bounds. It can be proven that no "classical" computer can search for an item in an unsorted list in better than O(n) time. This not only makes intuitive sense, it can be mathematically proven. However, there is an algorithm to do the same thing on QC's in O(sqrt(n)) time.

Some of the new high nitrogen explosives are well suited to this application.

It will be difficult to replace all of the colors produced by metals and other compounds used in fireworks- some of which are quite toxic (strontium, cadmium, arsenic, antimony, PVC plastic, etc.). It will also be much more expensive. But high nitrogen explosives and newer organic compounds have a lot to offer the field- including colors you can't get with the old standbys.

Some of the high nitrogen stuff I used to work with was pretty interesting. Lots of newer, potentially safer compounds are in the pipeline- mainly for military applications, but they can be bastardized to, er, recreational purposes.

http://www.nis.lanl.gov/projects/robot//

I've been working on a project for a few months now, utilizing parts from old drives. I'm time deficient of late, but I'm hoping when I finish a current work project, I'll have more time.
All you tinkerer nerds out there, if you haven't looked into BEAM robotics, look into it. You can utilize a good deal of junk electronics.

Very interesting! It almost sounds (remotely anyway) like magnetized target fusion (MTF). Everything old is new again I suppose...

Finally, you state that you do not wish to limit the ability of consumers to record over-the-air radio broadcasts. Instead, you apparently want to force them to buy what they have received for free since Fleming and Marconi first made it possible for consumers to hear news and music over the public airwaves.

As you know, we have long been concerned about content owners seeking to change the "play" button on our devices to a "pay" button. At least you have addressed the semantics by suggesting new devices come equipped with a "buy" button.*

*Excerpt of letter to Cary H. Sherman, President, RIAA, from Gary Shapiro, President & CEO, Consumer Electronics Association, 4/15/04, responding to a fax, by Cary H. Sherman to Gary Shapiro, at 14:30, 4/14/04, with this attempt at greasing the wheels:

We also point out that a lack of content protection will forever preclude a myriad of new business models that could [insert hush money offer here] benefit your members [end insert of hush money offer here] as well as other interested parties. For example, device manufacturers could provide "buy buttons" that would offer consumers the ability to quickly and easily purchase music that they hear on the radio. Indeed, iBiquity has said that it would like to offer to consumers, for a fee [as opposed to free], on demand weather and traffic reports. The same opportunity could and should exist for music, the bread and butter of radio broadcasts.

Sources

), they'll need control over the BIOS. That's why Microsoft is pushing so hard in this area, and why the Linux community must push back just as hard, and get the Linux BIOS.

5 second boot times were promised with a Linux BIOS. I'm still waiting. For the BIOS, and for my computer to boot up.

Another recen

Here's a full pre-print of the article.

[TMB]

Perhaps the best question to ask in this situation is why these labs are still using removable data storage devices to store sensitive information.

As a quick reality check, visit LANL's ASC site to convince yourself that (1) there's no way that they are carrying all that data around on floppies and (2) that given the scope of the computational effort, there are probably some operations that exceed the capability of a Javastation, XTerminal, or diskless Linux box.

TRANSIMS is an agent based model with each human accounted for in the system based on census info. So it is complex enough to handle any traffic pattern. It runs on LANL's ASCI Blue SGI cluster.

The TRANSIMS project has been around since at least 2000 so this latest project isn't the first to analyze traffic.

(L. alumen: alum) The ancient Greeks and Romans used alum as an astringent and as a mordant in dyeing. In 1761 de Morveau proposed the name alumine for the base in alum, and Lavoisier, in 1787, thought this to be the oxide of a still undiscovered metal.

Wohler is generally credited with having isolated the metal in 1827, although an impure form was prepared by Oersted two years earlier. In 1807, Davy proposed the name aluminum for the metal, undiscovered at that time, and later agreed to change it to aluminum. Shortly thereafter, the name aluminium was adopted to conform with the "ium" ending of most elements, and this spelling is now in use elsewhere in the world.

Aluminium was also the accepted spelling in the U.S. until 1925, at which time the American Chemical Society officially decided to use the name aluminum thereafter in their publications.

So it changed in the 1920's. I guess all that Charlston'ing went to people's heads.

Couldn't this be placed in a more, you know, useful location? I hate to offend the New Mexicanites but why is this not in San Diego where I live?

The albuquerque-Santa Fe metro area not only has two national laboratories Sandia and LANL, but it has a huge INTEL factory in Rio Rancho.

New Mexico also has, despite being named The dumbest state in the US(I dislike harvard...almost as much as I dislike MIT...) New Mexico has the highest concentration of Ph.D.'s in the United States today.

Get real...New Mexico is WAYYYY more relevant than San Diego...besides the Chargers and Aztecs both suck.

Van los lobos, baten a aztecas!

BTW it is New Mexicans, NOT New Mexicanites

Just some examples of how bio very much is open already..

In biotech software, there's lots of open source. BioJava, BLAST.. etc.

As for what they're talking about, e.g. databases.. Most data already IS open. The human genome, protein structures and sequences.

A cockroach hmm? Try BEAM robotics, perhaps the unibug? Analog only, yet still much better than most digital bots. I'd say that's a good enough new approach to start out with.

Seems like a prime opportunity for a nuclear powered rover!

The chemistry is pretty simple: C + H2O + O2 -> CO + CO2 + H2 in various ratios. In the applications I've studied, the carbon was in the form of coal or petroleum coke and the gas was subsequently scrubbed of the H2S produced by reduction of pyrites.

If you are trying to produce hydrogen, your carbon emissions will be 100% of the input carbon. However, the carbon comes out as CO2 under high pressure during the scrubbing step; this is easy to capture and can be disposed of in various ways (e.g. into wells) at relatively low cost.

If you're trying to produce lighter fractions from bitumen, I'll bet that the hydrogen is used to turn long-chain hydrocarbons (C50?) into C6-C10 chains. This will require roughly 1 molecule of hydrogen per 6-10 atoms of carbon in the input; if you can produce 2 moles of H2 from 1 mole of carbon into the gasifier, the losses should be relatively low. You'd have the further benefit of having the byproduct heat of the gasifier and oxygen plant to heat the water used to recover the bitumen from the sand (or you could even dissolve the bitumen off the sand using supercritical CO2 as a solvent; you'd have plenty of CO2 available to run the process).

As I am not a petrochemical engineer, I could be missing something about this that makes it completely unrealistic as an approach to the problem. But it looks good from a distance.

While the "Many Worlds" interpretation could explain this phenomenom it cannot (to my knowledge) explain other QM phenomena which seem to require faster-than-light communication or at least "pre-knowledge" of what is going to happen.

What about Richard Feynman ...

See the references to Feynman's work in Deutch's 1985 paper Quantum theory, the Church-Turing principle and the universal quantum computer :

Feynman (1982) went one step closer to a true quantum computer with his 'universal quantum simulator' [.... although] it is not a computing machine in the sense of this article.

jfern wrote:

... or Peter Shor?

There are many purpose-built supercomputers coming up (like Sandia's Red Storm) that use custom yet pricy interconnects that end up smoking anything Quadrics can put together. Anytime your interconnect relies on a PCI-type bus, you take a latency penalty on each end. Real supercomputers access memory on other nodes directly, not through a generic shared bus to a fancy network card.

Read some of this if you're bored, it goes through Sandia's entire thought process. http://www.lanl.gov/orgs/ccn/salishan2003/pdf/camp .pdf

This has been released very recently - it's based on PbSe crystals instead - at Los Alamos but also through University of California.

The numbers come from NASA's Wilkinson Microwave Anisotropy Probe (WMAP) which measured fluctuations in the Cosmic Microwave background (afterglow of the Big Bang). There's a good review of their results in hep-ph/0308251 accessible from the LANL preprint server though it might be a bit technical for most.

That the sensor has never detected something doesn't tell you that it's working or not working - or am I am missing something here?

Yes, you're missing something. :-)

The statement "the sensor has never detected something" is patently false. Figure 1 of the paper shows all of their detections - and there are lots of them! WIMPs aren't the only things that interact with Germanium. ;-) However, once you exclude all of the events which are consistent with being cosmic-ray produced interactions with the shielding, you get Figure 4... all of the detections in the red region (which is where the WIMPs would show up) are gone.

So the detector works great and detects lots of things! But no WIMPs yet.

[TMB]

Guys, it's a thermoacoustic Stirling. Stirling engines can do a variety of things. You can supply heat, and they will supply motion (usually a reciprocating motion, which works nicely when connected with a linear alternator). Or, you can supply motion (reciprocating), and they will move heat (i.e. usually cooling something). Sunpower has been making cryocoolers based on them for years. And yes, they even played with a module which would attach to a CPU and supercool it.

Hit Google and look for information on Stirling engines. Then, reduce the number of moving parts. Instead of a power piston and a displacer, they use sound waves as the displacer and the speaker as the power piston.

There are already companies at work trying to commercialize this technology. The guys at Purdue are re-inventing the wheel. Check the first link about thermoacousting Stirling engines, and you'll see they've been working on this for some time.

According to Google it's a lot of things, including a 3D image archive, a comic strip, an Apache project for service and component management or a Beowulf cluster.
However, there's no Microsoft stuff on the first ten hits.

I was interested in how they generated the entangled photon pairs, so I googled and came up with an interesting paper that touches on the subject:

"The entangled photon pairs created by Kwiat's team are produced using two thin, nonlinear optical crystals to split the "parent" photons from a laser into entangled "daughter" photons. In previous research at Los Alamos, these entangled photons have been used for quantum cryptography to create unbreakable cryptographic keys that can be used to lock or unlock encrypted messages.

Decoherence is a problem in quantum systems because the fragile quantum superpositions of entangled states are destroyed by unwanted coupling to the environment through which the photons are passing. Decoherence in Kwiat's system is intentionally created by passing the entangled photons through a roughly 10 millimeter piece of quartz. This optical environment produces a collective decoherence in the photons where one particular entangled photon state is, as predicted by quantum theory, essentially decoherence-free. These photons could serve as the basis of information carriers for quantum communications."

Brazil Nuts are naturally high in barium (0.3% by weight) and radium -- making it one of the most radioactive foods.

I wonder if plants can be used to extract waste pharmaceuticals out of the ground, too, such as destruxol and THC.

It's a pretty elegant theory, and explains a lot of anamolies that don't really satisfy current physical models.

"How are you going to adjust your encryption when quantum computers will make most encryption schemes obsolete? "

Why FUD my friend? This just isn't true!

The truth about quantum cryptography is that RSA and DH will be destroyed by quantum cryptography. This is due to the work of Shor who famously proved that you could factor in cubic time.

This sounds bad but we've already had good success in performing quantum key exchanges (that are unbreakable in a theoretical sense).

What does this mean for symmetric cryptography such as AES? Well, Quantum Computers that deploy Grover's algorithm, can search unordered lists in under sqrt(n) operations. A normal computer does this in an average of n/2 steps. The key space of a cipher is an unordered list so we'd only have to double our keysizes to avoid the "Grover attack".

Clearly more research is needed but the quantum future is bright as far as cryptography is concerned.

Simon.

Hmmm... Sandia and several other US government labs seem to think different. Exactly *what* class of computations can a linux cluster not handle?

UC Irvine

University of Cyprus

Linux supercomputer for Los Alamos

AMD Tapped for Gov. Linux Clusters

Installing, Running and Maintaining Large Linux Clusters at CERN

And more....

Perhaps taking PIXAR's high profile into account, Terry concedes that "The commercial environment is still best served by Linux clusters" at the end of the article.

As one of Cray's most prominent customers, Los Alamos National Labs (LANL) implicitly makes a better argument. They've long advocated clusters for massive computing purposes, as evidenced by the Avalon, Loki, et al.

It's also worth mentioning that Loki won the Gordon Bell Price/Performance Prize in 1997.

And let's not forget the Lawrence Livermore National Laboratory's Beowulf cluster(s).

Fun side note: I've never gotten to see these clusters in RL, but I did get to see Sandia National Labs' Paragon and some other boxen when I was there in 1995 as part of the AiS Challenge.

Perhaps taking PIXAR's high profile into account, Terry concedes that "The commercial environment is still best served by Linux clusters" at the end of the article.

As one of Cray's most prominent customers, Los Alamos National Labs (LANL) implicitly makes a better argument. They've long advocated clusters for massive computing purposes, as evidenced by the Avalon, Loki, et al.

It's also worth mentioning that Loki won the Gordon Bell Price/Performance Prize in 1997.

And let's not forget the Lawrence Livermore National Laboratory's Beowulf cluster(s).

Fun side note: I've never gotten to see these clusters in RL, but I did get to see Sandia National Labs' Paragon and some other boxen when I was there in 1995 as part of the AiS Challenge.

doesn't this CTO of cray remind you of someone?
"There IS no Linux in high-performance clusters."

"There IS no Americans in Iraq."

OMG! It's the former Iraqi mis-Informed-ation minister!

Especially when 2004 has been dubbed the year of the penguin, it's wreckless to claim that Linux can't be used in HPC's.
Hell, just look at the current top500 list. There's no Cray in the top 10 but there are two Linux based clusters there (and one based on OSX [FreeBSB based]).

Here's a few:
NCSA's IA32 Linux cluster
NCSA's IA32 Linux cluster
Space Simulator Clust at Los Alamos (SS51G based; makes me proud as I have a SS51G too)
Beowulf - used in many Linux clustering projects
Linux clusters at Los Alamos (they seem to have more than one)
Virginia Tech's Supercomputer X

Although there are techniques for safely testing nuclear rockets on earth and some nuclear rocket testing going on, space researchers are still a bit too timid to go ahead with a nuclear rocket program. The advantages of nuclear rockets over conventional rockets include massive reduction in rocket mass and increase in specific impulse. Testing of nuclear rockets on the Moon (which is already highly radioactive due to constant bombardment from the solar wind) will prepare us for a fast journey to Mars, i.e., weeks not years. Then we can make multiple trips and build massive ships for journeys to the outer planets.

On a related note, LANL has announced a plan to begin above-ground weapon testing in the Salt Lake City area...

So in my opinion there are 3 good websites on the internet:
E-Print Archives
Mathworld and Scienceworld
Federation of American Scientists

Of the three, 2 are distinctly not for profit, but rather so that scientists can get some work done again and who know's why wolfram put mathworld and scienceworld online. As far as more liberal arts stuff, the only online thing I know of is jstor.org and I think that might require paying for, but my university pays for it if it does. I found all those sites very useful and suggest that you check them out if you haven't already done so.

-Scott

Don't confuse LLNL with LANL. Las Alamos National Labs is where Linux Networx i think installed the Pink Beowulf clusters.

I think mentioning the war on terror is legit, because muon radiography can also be used to detect nukes in cargo containers