Slashdot Mirror

I would be far more interested in taking advantage of all the CPU cycles that run all over at Businesses.

Condor.

http://webstreamer2.doit.wisc.edu/titan_arum/galle ry01/pages/4_full.htm

The boy in the back looks like someone's told him the truth about mammal decomposition.

The graduate students explaining the environment the flower is found in, as well as the amazing way in which it attracts and traps flies/carrion beetles to reproduce and the 10-foot tall tree-like leaf (!) make me truly appreciate how amazing the spectrum of life we have here on Earth is. It's certainly above my head -- that's why I can stick to relatively simple things, like working on a plasma confinement device. ;)

For those of you who can't stand your piddly little 384 kbps stream

...and are feeling unfulfilled by your 1.5 Mbps stream

...Don't worry. They have set up an experimental 30 Mbps DV multicast stream , complete with it's own free DVTS player so you can catch all this incredibly FAST paced action, as it happens!

For those of you who can't stand your piddly little 384 kbps stream

...and are feeling unfulfilled by your 1.5 Mbps stream

...Don't worry. They have set up an experimental 30 Mbps DV multicast stream , complete with it's own free DVTS player so you can catch all this incredibly FAST paced action, as it happens!

For those of you who can't stand your piddly little 384 kbps stream

...and are feeling unfulfilled by your 1.5 Mbps stream

...Don't worry. They have set up an experimental 30 Mbps DV multicast stream , complete with it's own free DVTS player so you can catch all this incredibly FAST paced action, as it happens!

But wow, check out the mullet on the guy here! :)

photos and description

http://mrsec.wisc.edu/Edetc/nanolab/gold/text.html

The University of Wisconsin.

Text of the above article:

The University of Wisconsin - Madison has deployed 35 5.6TB Xserve RAID storage arrays in a single research installation as part of an ongoing scientific computing initiative.

The Grid Laboratory of Wisconsin (GLOW), a partnership between several research departments at the University of Wisconsin, have installed almost 200TB, or 200,000GB, of Xserve RAID arrays. As a comparison, 200TB of storage is enough to hold 2.75 years of high definition video, 25,000 full length DVD movies, 323,000 CDs, 20 printed collections of the Library of Congress, or over 1000 Wikipedias.

The GLOW storage installation is physically split between the departments of Computer Sciences and High Energy Physics. Each Xserve RAID is attached to a dedicated Linux node running Fedora Core 3 via an Apple Fibre Channel PCI-X Card and is either directly accessed via various mechanisms, such as over the network via gigabit ethernet, or aggregated using tools such as dCache.

The storage is primarily used to act as a holding area for large amounts of data from experiments such as the Compact Muon Solenoid (CMS) and ATLAS experiments at the Large Hadron Collider at CERN.

Aside from the GLOW initiative, the university also has Xserve RAID storage systems in use in other areas as well.

Full disclosure: I am the administrator of alienraid.org and am affiliated with the University of Wisconsin.

The University of Wisconsin has deployed 200 TB of storage for support of similar types of experiments as part of the Grid Laboratory of Wisconsin.

Brief article, with pictures:

University of Wisconsin deploys nearly 200TB of Xserve RAID storage (Google cache)

The storage is used for, among other things, particle physics simulations in support of research projects at sites such as the Large Hadron Collider at CERN. More information on GLOW and its initiatives can be found here.

Text of the above article:

The University of Wisconsin - Madison has deployed 35 5.6TB Xserve RAID storage arrays in a single research installation as part of an ongoing scientific computing initiative.

The Grid Laboratory of Wisconsin (GLOW), a partnership between several research departments at the University of Wisconsin, have installed almost 200TB, or 200,000GB, of Xserve RAID arrays. As a comparison, 200TB of storage is enough to hold 2.75 years of high definition video, 25,000 full length DVD movies, 323,000 CDs, 20 printed collections of the Library of Congress, or over 1000 Wikipedias.

The GLOW storage installation is physically split between the departments of Computer Sciences and High Energy Physics. Each Xserve RAID is attached to a dedicated Linux node running Fedora Core 3 via an Apple Fibre Channel PCI-X Card and is either directly accessed via various mechanisms, such as over the network via gigabit ethernet, or aggregated using tools such as dCache.

The storage is primarily used to act as a holding area for large amounts of data from experiments such as the Compact Muon Solenoid (CMS) and ATLAS experiments at the Large Hadron Collider at CERN.

Aside from the GLOW initiative, the university also has Xserve RAID storage systems in use in other areas as well.

Full disclosure: I am the administrator of alienraid.org and am affiliated with the University of Wisconsin.

The University of Wisconsin has deployed 200 TB of storage for support of similar types of experiments as part of the Grid Laboratory of Wisconsin.

Brief article, with pictures:

University of Wisconsin deploys nearly 200TB of Xserve RAID storage (Google cache)

The storage is used for, among other things, particle physics simulations in support of research projects at sites such as the Large Hadron Collider at CERN. More information on GLOW and its initiatives can be found here.

Text of the above article:

The University of Wisconsin - Madison has deployed 35 5.6TB Xserve RAID storage arrays in a single research installation as part of an ongoing scientific computing initiative.

The Grid Laboratory of Wisconsin (GLOW), a partnership between several research departments at the University of Wisconsin, have installed almost 200TB, or 200,000GB, of Xserve RAID arrays. As a comparison, 200TB of storage is enough to hold 2.75 years of high definition video, 25,000 full length DVD movies, 323,000 CDs, 20 printed collections of the Library of Congress, or over 1000 Wikipedias.

The GLOW storage installation is physically split between the departments of Computer Sciences and High Energy Physics. Each Xserve RAID is attached to a dedicated Linux node running Fedora Core 3 via an Apple Fibre Channel PCI-X Card and is either directly accessed via various mechanisms, such as over the network via gigabit ethernet, or aggregated using tools such as dCache.

The storage is primarily used to act as a holding area for large amounts of data from experiments such as the Compact Muon Solenoid (CMS) and ATLAS experiments at the Large Hadron Collider at CERN.

Aside from the GLOW initiative, the university also has Xserve RAID storage systems in use in other areas as well.

Full disclosure: I am the administrator of alienraid.org and am affiliated with the University of Wisconsin.

The University of Wisconsin has deployed 200 TB of storage for support of similar types of experiments as part of the Grid Laboratory of Wisconsin.

Brief article, with pictures:

University of Wisconsin deploys nearly 200TB of Xserve RAID storage (Google cache)

The storage is used for, among other things, particle physics simulations in support of research projects at sites such as the Large Hadron Collider at CERN. More information on GLOW and its initiatives can be found here.

Text of the above article:

The University of Wisconsin - Madison has deployed 35 5.6TB Xserve RAID storage arrays in a single research installation as part of an ongoing scientific computing initiative.

The Grid Laboratory of Wisconsin (GLOW), a partnership between several research departments at the University of Wisconsin, have installed almost 200TB, or 200,000GB, of Xserve RAID arrays. As a comparison, 200TB of storage is enough to hold 2.75 years of high definition video, 25,000 full length DVD movies, 323,000 CDs, 20 printed collections of the Library of Congress, or over 1000 Wikipedias.

The GLOW storage installation is physically split between the departments of Computer Sciences and High Energy Physics. Each Xserve RAID is attached to a dedicated Linux node running Fedora Core 3 via an Apple Fibre Channel PCI-X Card and is either directly accessed via various mechanisms, such as over the network via gigabit ethernet, or aggregated using tools such as dCache.

The storage is primarily used to act as a holding area for large amounts of data from experiments such as the Compact Muon Solenoid (CMS) and ATLAS experiments at the Large Hadron Collider at CERN.

Aside from the GLOW initiative, the university also has Xserve RAID storage systems in use in other areas as well.

Full disclosure: I am the administrator of alienraid.org and am affiliated with the University of Wisconsin.

FYI, I'm not an American, and the page about Cuban human rights abuses which I linked to is a European/international human rights group. So I don't think your generalization about other countries is particularly relevant. It's true that many Europeans seem particularly insensitive to issues of political repression, but that's perhaps why they keep getting into trouble along those lines.

I have lived for a number of years in America, though, and my experience is that ordinary citizens there aren't afraid of their government(s) in the same way that they are in Cuba. Any comparison in terms of repressiveness between the two countries is largely silly, even despite the current overreaction to terrorism. There's a spectrum of human rights issues, and no country on Earth is perfect, and particularly no English-speaking country (if there was one, I would live there). Smooshing the spectrum to make all violations equal to each other is only useful as an extremist rhetorical tactic.

I don't buy the pot smoker comparison, either: show me the pot smoker (not dealer) who is in jail for one to four years for mere possession of single-person quantities. At least such a person committed a crime, though, even if it shouldn't be a crime: they get due process under the law, unlike Cubans imprisoned for pre-crimes. You can't have due process when you don't know what actions might trigger your arrest and imprisonment.

Regarding US imprisonment rates, that seems to be largely a racial thing. Amazing how long the legacy of slavery has lasted. So yes, it can suck to be black in America. In Cuba, it sucks to be Cuban.

FYI, I'm not an American, and the page about Cuban human rights abuses which I linked to is a European/international human rights group. So I don't think your generalization about other countries is particularly relevant. It's true that many Europeans seem particularly insensitive to issues of political repression, but that's perhaps why they keep getting into trouble along those lines.

I have lived for a number of years in America, though, and my experience is that ordinary citizens there aren't afraid of their government(s) in the same way that they are in Cuba. Any comparison in terms of repressiveness between the two countries is largely silly, even despite the current overreaction to terrorism. There's a spectrum of human rights issues, and no country on Earth is perfect, and particularly no English-speaking country (if there was one, I would live there). Smooshing the spectrum to make all violations equal to each other is only useful as an extremist rhetorical tactic.

I don't buy the pot smoker comparison, either: show me the pot smoker (not dealer) who is in jail for one to four years for mere possession of single-person quantities. At least such a person committed a crime, though, even if it shouldn't be a crime: they get due process under the law, unlike Cubans imprisoned for pre-crimes. You can't have due process when you don't know what actions might trigger your arrest and imprisonment.

Regarding US imprisonment rates, that seems to be largely a racial thing. Amazing how long the legacy of slavery has lasted. So yes, it can suck to be black in America. In Cuba, it sucks to be Cuban.

Over 90% of the helium ore mined in the world comes form the helium mine at Amarillo.

There is no such thing as "helium ore".

"[...] helium is produced chiefly in Texas and Kansas, where it occurs as a minor constituent of the natural gas produced with oil there."

Learn more about Helium.

I have [...] no vested interest of any kind[.]

There's also this one. Anyone who thinks that snake was just falling needs their eyes checked. Even an experienced skydiver would be impressed with that.

That's awesome! That snake really looks good, until this jump. Oops?

Agreed, until I saw someone above link to this one which I didn't see on the original site. That's definitely impressive as it seems to almost do better than a 45 degree angle.

Pretty true from what I can see, except this one plummets at a pretty good angle.

Mirror of the flying snake "takeoff" and "gliding" videos.

(The author has also stated "You may use this video footage for non-publishing purposes without permission" on his web page. Not sure how much load it can handle, so I figured I'd mirror it. The author's page has more information and commentary; these are just the raw videos. They're very small, so there's a good chance his server could have handled it just fine.)

1) He says barriers to entry of startups is low. Wrong. Barriers to entry of web services is low; most other startups still require huge infusions of capital

2) Low barriers to entry also means there is going to be hundreds of other "undergrads" trying to sell the same idea. This means your chances of eventual payback are much smaller!

3) Why should bigger companies buy startups when they can just partner with them or outsource company services to them?

4) Yeah, starting a web based startup doesn't cost significantly more than just being a slacker. But if you haven't noticed, 99% of us can't afford to just set around and be a slacker either! SOMEBODY has got to be paying your food and rent. Apparently Mr. Graham thinks most students graduate with tens of thousands of dollars in the bank and can afford to not have any income for several years. I've got about $130,000 in student loans that say otherwise...

Try this one on for size: My Alma Mater currently costs $5866.24 per year for tuition (GO BADGERS!). Live at home or get a cheap-ass apartment with a few roommates for $100/mo. Eat Ramen. Your college costs are about $10,000/year when you get done with books and stuff.

Now here's where it gets tricky. Work June through August, and winter break as a temp making $12/hr. That's $8,000. Now work in the campus library 10 hrs/wk at $9/hr which is $3000/school yr. Your total income for the year is $11,000, and you should pay no tax with all the education tax credits.

In my case, I worked as a waiter for $20/hr, so I didn't have to eat Ramen... I just stole food from the restaurant. Show a little creativity!

Anyhow, that is the correct way to pay for college if your parents can't/won't pay for you. Taking out $130,000 in student loans is NOT the correct way to pay for college. I have no sympathy for people who do that.

What an original idea! You mean like this? What is with Europe? Mars Express, A380 and now this? Tone it down guys.

That's very similar to Europe. They have a higher percentage of broad band usage than the U.S. They also have a higher population density.

Even more interesting but off topic. Compare the population density map of the US to the 2004 election results.

Look at a population map of china. Now look at one of the US. Half of china has a population density of less than 2.5 people per square mile, and the other half has more than 500. Getting a large section of the country wired very quickly is pretty easy. The US population, on the other hand, is spread really thin. So it's not surprising that China could overtake the US quickly.

PDF is an open format. Ther are several non_Adobe PDF readers, eg GSview.

Simultaneous MPEG2 decoding, as shown here, is what computer architects call an embarrassingly-parallel problem. The easiest way to speed it up is just add more processors - with 8 processing units, the Cell is a great fit.

However, the really interesting problems are the ones that don't scale linearly in performance with the number of processors - these are the tasks for which the Cell processor will probably be running with 7 idle units and 1 active. These are also the tasks where we need actually new architectures; supercomputers like BlueGene will tear their way through extremely parallel problems.

One very cool approach to handling less parallel workloads (or even "sequential" workloads - like the majority of programs people usually run on PCs) is speculative threading - taking a sequential program, breaking it up into chunks, and running those chunks in parallel. Of course, when you do this, you have to make sure that the later work doesn't depend on the earlier operations, and check for violations of "sequential execution semantics" (programs expecting sequential execution semantics are ones that expect their instructions to execute in order - basically any program you'd write today). The Stanford Hydra project is an example that uses this technique; Wisconsin Multiscalar Group takes an approach that requires modified binaries to do something similar.

One thing people fail to mention when they talk about the supposedly-amazing performance of the Cell processor is its floating point precision: first, it only attains it's >200GFLOPS with single precision numbers (not accurate enough for many scientific applications), and second, it doesn't follow IEEE754 rounding requirements. The rounding policy in IEEE754 floats is specifically designed so that as you perform more and more calculations, the error doesn't grow rapidly. Cutting corners lets you calculate faster but even less accurate numbers. Basically, to get the high FLOPS ratings, Cell sacrifices precision in both the number of bits used, and the accuracy of the data in those bits.

It's at the end of this large PDF.

http://www.ssc.wisc.edu/cde/cdewp/98-07.pdf

The authors also admit that on April 1st they pointed the neutrons at the South Pole and Amanda.

Oh?

Hear, hear.

I've been quite surprised at the influx of "odd" observations over the past few years; I certainly wasn't expecting local pancake structures.

You raise a pretty good point, though, on the structure of disks, large and small, in the first place.

Plasma physicists jump up and down that the in-vogue theories treat large-scale magnetic fields and currents as non-existent, as though charge must cancel out on the large scale, therefore it has no effect. Sometimes, they make a good point - some of the disk systems do resemble dynamos.

Some of the papers I've read in passing on "push" gravity theories estimate that the force of gravity is proportional to 1/d**2 locally, but trends to 1/d on the outsides of the galaxy. Otherwise, there's a lot of unseen matter there (and we haven't seen anything resembling the high-velocity clouds gathering on the edges of the galaxy)... or, alternately, we're ignoring a dynamo effect.

Or... etc. (Assuming we stop before postulating that angels sit on the edge fanning galaxies with their wings ;)

It's the bank of poorly-explained pieces that will lead us to our next big theoretical breakthrough (or revolution) - but it takes some special vigilance to keep track of what hasn't actually been explained properly, and what's been merely papered over.

Too many tweaks. They should have realized something was wrong sometime between inflation theory, and dark-energy-requiring ever-increasing-acceleration theory. Plenty of duct tape on things already :)

By the way, speaking of aether... ;)

I can understand the establishment position somewhat... it's either duct tape or anarchy. There's got to be a standard to measure against, but if the explanations start stretching thin, they need an exit strategy.

If that day comes, they will need to exit to something, though. What's out there that can explain the pancakes at multiple scales of the universe and other phenomena as well?

Perhaps they need to take a page out of other research and development, and apportion some funds to "blue sky" research.

The biggest dividends will come from research that's reviewed for logic, self-consistency and explanation of phenomena without regard to how well it fits into prior patterns. Pro-Ams and people in fields with more easily measureable results (applied sciences, for one) realize these benefits, but being in a field where so many assumptions have to be made to interpret the results in the first place make this next to impossible for the theoreticians to condone dissent.

Everybody's MMV :)

-- Ritchie

Well, if anyone needs the visual proof that carbon nanotubes are very electrically conductive, here's a video.

And just in case anyone wants to know what exactly, a 5,5 armchair nanotube looks like, there are some images of models here.

Thankfully, as you alluded, mutlicast capability lives on in Internet2:

http://multicast.internet2.edu/

At the University of Wisconsin, our new 10Gbps ethernet backbone and all associated equipment in a major network upgrade initiative supports multicast to the desktop. We're operating an IP-based television distribution system exclusively via multicast distribution (using locally scoped addresses, so it's only available internally).

So we can still go to 224.2.231.45, and get a live stream of NASA TV from the University of Oregon.

For the uninitiated, multicast essentially allows any number of clients to "listen" to the same stream: multicast-aware network equipment just handles when a network gets traffic. If a user on the University of Wisconsin campus decides to watch the broadcast from the University of Oregon, one stream's worth of bandwidth will enter our network. If a hundred - or a thousand - people decide to watch it, it's still that same one stream's worth of bandwidth coming in, that everyone else is simply "listening" to. So for each network segment, whether you're looking at an individual subnet or in a whole-internet sense, there is either:

- 0 streams
- 1 or more streams, but all with the equivalent network usage of 1 stream

It's really a fantastic way of distributing video. Not only is there no additional load beyond the one stream on the network, but there is also *only the load of one stream* on the server.

If multicast were enabled on the internet-at-large, individual people really could distribute video to the world: all they'd need is essentially enough bandwidth to distribute one stream, and one, or one million, could listen in.

(And yes, there are ways this can break down, but I'm just trying to give a simplified explanation here.)

My first year at the UW(1997), the CS school had four solaris labs. They are all linux now.

http://www.cs.wisc.edu/~coates/video/Redefined%20- %20Nintendo%20(Video).mpg

The system reads a high-level description of a document similar in style to LaTeX and produces a PostScript file which can be printed on most laser printers and graphic display devices. Plain text output is also available, PDF output is not supported anymore.

The system reads a high-level description of a document similar in style to LaTeX and produces a PostScript file which can be printed on most laser printers and graphic display devices. Plain text output is also available, PDF output is not supported anymore.

Actually, there might be a way to get a little bit of warning, depending on the source of the gamma ray burst.

Photons (gamma rays) take a long time to get out of a star. But neutrinos, because of their physical properties, pass right through most of the star. Most nuclear reactions that generate photons also generate neutrinos. They're just very hard to detect (because of that same physical property).

Well, I'm working on a neutrino detector at the South Pole right now. http://icecube.wisc.edu/

It could, when it's complete, pinpoint the source of the neutrinos. Given the energy level of the neutrinos and the sudden, large burst of them, a whole lot of scientists are going to be woken up - and I mean that literally.

An earlier version of the project, AMANDA http://amanda.wisc.edu/, already has a supernova detector. It hasn't gone off yet, but when it does it will start a sequence of events that ultimately steers a lot of telescopes to point at that supernova.

Actually, there might be a way to get a little bit of warning, depending on the source of the gamma ray burst.

Photons (gamma rays) take a long time to get out of a star. But neutrinos, because of their physical properties, pass right through most of the star. Most nuclear reactions that generate photons also generate neutrinos. They're just very hard to detect (because of that same physical property).

Well, I'm working on a neutrino detector at the South Pole right now. http://icecube.wisc.edu/

It could, when it's complete, pinpoint the source of the neutrinos. Given the energy level of the neutrinos and the sudden, large burst of them, a whole lot of scientists are going to be woken up - and I mean that literally.

An earlier version of the project, AMANDA http://amanda.wisc.edu/, already has a supernova detector. It hasn't gone off yet, but when it does it will start a sequence of events that ultimately steers a lot of telescopes to point at that supernova.

Hear, hear.

I've been quite surprised at the influx of "odd" observations over the past few years; I certainly wasn't expecting local pancake structures.

You raise a pretty good point, though, on the structure of disks, large and small, in the first place.

Plasma physicists jump up and down that the in-vogue theories treat large-scale magnetic fields and currents as non-existent, as though charge must cancel out on the large scale, therefore it has no effect. Sometimes, they make a good point - some of the disk systems do resemble dynamos.

Some of the papers I've read in passing on "push" gravity theories estimate that the force of gravity is proportional to 1/d**2 locally, but trends to 1/d on the outsides of the galaxy. Otherwise, there's a lot of unseen matter there (and we haven't seen anything resembling the high-velocity clouds gathering on the edges of the galaxy)... or, alternately, we're ignoring a dynamo effect.

Or... etc. (Assuming we stop before postulating that angels sit on the edge fanning galaxies with their wings ;)

It's the bank of poorly-explained pieces that will lead us to our next big theoretical breakthrough (or revolution) - but it takes some special vigilance to keep track of what hasn't actually been explained properly, and what's been merely papered over.

Too many tweaks. They should have realized something was wrong sometime between inflation theory, and dark-energy-requiring ever-increasing-acceleration theory. Plenty of duct tape on things already :)

By the way, speaking of aether... ;)

I can understand the establishment position somewhat... it's either duct tape or anarchy. There's got to be a standard to measure against, but if the explanations start stretching thin, they need an exit strategy.

If that day comes, they will need to exit to something, though. What's out there that can explain the pancakes at multiple scales of the universe and other phenomena as well?

Perhaps they need to take a page out of other research and development, and apportion some funds to "blue sky" research.

The biggest dividends will come from research that's reviewed for logic, self-consistency and explanation of phenomena without regard to how well it fits into prior patterns. Pro-Ams and people in fields with more easily measureable results (applied sciences, for one) realize these benefits, but being in a field where so many assumptions have to be made to interpret the results in the first place make this next to impossible for the theoreticians to condone dissent.

Everybody's MMV :)

-- Ritchie

Get Comcast to do this in a few major cities and you've made a serious dent in the problem. Comcast would save big money by decreasing traffic on their backbones, and NASA would obviously save a lot too. Now the only question is how to make streaming media proxy-cache friendly.

Hmm, apparently it has been done. I wonder if the big ISPs do this?

Hey buddy, how about providing a link next time?

Found via Google Search.

There is a package called Condor which runs as a job-schedular under windows or linux. It can either run on dedicated machines or as a seti-at-home screensaver type idea to use idle machines. Works very nicely and can really harvest a lot of idle PC's across companies or campuses...

And of course since you come from the Wisconsin Department of Transportation (per your http://das.doit.wisc.edu/ link), you obviously offer an unbiased opinion.

For Windows just install the free Ghostscript plus CuteWriter.

Hey Matt! The file named "matt_poster_small.jpg" is a 6 Mb, 4400 x 3300 pixels image. Not exactly a thumbnail. Besides, I have no 42 inches display, you insensitive clod!

Hey Matt! The file named "matt_poster_small.jpg" is a 6 Mb, 4400 x 3300 pixels image. Not exactly a thumbnail. Besides, I have no 42 inches display, you insensitive clod!

Hey Matt! The file named "matt_poster_small.jpg" is a 6 Mb, 4400 x 3300 pixels image. Not exactly a thumbnail. Besides, I have no 42 inches display, you insensitive clod!

Infrared pictures are just not as pretty as those made in visible light.

I beg to differ. And I'm an optical astrnomer.

Nearly all PR images released by professional astronomers (and many by amateurs)--especially NASA--are not "true color". Nearly every multi-color image you see is from multiple actual observations that are combined in an image processing program. How "pretty" you want to make them is partly dependent on how much time, experience, and (frankly) money you have to spend on them. Keep in mind that "prettifying" pictures does not necessarily make them better for the science.

And all pretty cosmic phenomena Spitzer can see are also observable by the Hubble telescope.

Absolutely false.

Interstellar dust attenuates light, especially in the plane of the Milky Way. The survey project I linked to above penetrates much further than any visible light instrument can. And they only use 2 second integrations. Deeper, focused observations can detect distant gas and stars even better.