Slashdot Mirror

The idea is very old, and contrary to the article there are plenty of people offering similar services: http://news.softpedia.com/news/Rent-Your-Own-Supercomputer-for-2-77-per-Hour-82166.shtml, http://www.hoise.com/primeur/00/articles/weekly/AE-PR-04-00-20.html, http://www.linuxplanet.com/linuxplanet/reports/4590/2/, etc.

Is their offering cheaper? Unfortunately the article didn't tell us.

how would data parallelism negatively affect a test that is designed to measure a system's performance in supercomputing applications--a field which is dominated by problems which involve processing extremely large data sets?

if vector processors do in fact perform poorly on LINPACK benchmarks then that would mean LINPACK performance is not a good indicator of real-world performance, but that clearly isn't the case as vector processors consistently perform quite well in LINPACK suite measurements.

vector processing began in the field of supercomputing, which during the 1980's and 1990's were essentially the exclusive realm of vector processors. it wasn't until companies, to save money, started designing & building supercomputers using commodity processors (P4s, Opterons, etc.) that general-purpose scalar CPUs began to replace specialized vector processors in high-performance computing. but now companies like Cray and IBM are starting to realize that this change was a mistake.

even in commodity computing the momentum is shifting away from general-purpose scalar CPUs towards specialized vector coprocessors like GPUs, DSPs, array processors, stream processors, etc. when you're dealing with things like scientific modeling, economic modeling, engineering calculations, etc. you need to crunch large data sets using the same operation; this is best done in parallel using SIMD. using specialized vector processors (and instruction sets) you can run these applications far more efficiently than you could using a scalar processor running at much higher clock speeds. the only downside is that you lose the advantage of using commodity hardware that's cheap because of their high volume production. but if companies like Adobe start developing their applications to employ vector/stream coprocessors, then that will boost the adoption of these vector processors in the commodity computing market, which will increase production volume and lower manufacturing costs.

They use commercial CFD to model the flow around the swimmer. Been years now. This story? Ooooh. Now they're running exactly the same thing on a faster box.

The first transatlantic operation dates back from september 2001, with a dedicated high-speed link from France Telecom. See http://www.hoise.com/vmw/01/articles/vmw/LV-VM-10- 01-20.html. It was not a heart operation (the bladder ?), and it was not robotic (there were real surgeons at the other end). But I am not sure it will be widely used, because you need to have very strong guarantees about the time lag, that you can not have on the Internet.

Despite the revisionism in the phrase "Cray co-founder", he is an acknowledged HPC expert. See Burton Smith's bio at the Computing History museum. However see also the rather self-serving Seymour Cray Award he received during his tenure at Cray. Now what's a hot iron hardware guy doing joining Microsoft you might be thinking? Well take a look at his articles.

Meanwhile FPGAs have displaced DSPs, FFTs and are overtaking CPUs for embedded applications. There are even rumblesof FPGAs seriously impacting the HPC market. Times are a changin so I'm not surprised to see traditional CPU-based CS research being downsized in response to this paradigm shift. Perhaps we need to take VIVA seriously just as Cray, SGI, Starbridge Systems, SRC, Nallatech and others are doing.

In the 1980s, IBM opened representative offices in Beijing and Shanghai, followed in 1992 by establishment of the IBM China Company Limited, a wholly-owned subsidiary of IBM World Trade Corporation. The IBM China Research Laboratory was established in Beijing in 1995. Today, IBM China has offices in 11 cities and operates eight joint venture companies in China.
--PrimeUR

IBM built and operates a chip packaging plant in China (registration site), a Research Laboratory in China, and is eyeing upward of a 50 percent share of China's market for business computers. Even IBM mainframes are big in China

IBM is creating a chip ecosystem in China and expects that Asian manufacturers will represent the bulk of the new Power licensees

Dr. Thomas A. Furness III

This is really really old news. And I'm not talking about a simulation either. Check out this page and scroll down to September 20, 2001. Here's another link.

Although there is more work to be done with the AcceleGlove, Hernandez-Rebollar is not sure if he will have the necessary financial support to continue his research after his dissertation.

The home page at Cray for the Cascade project.

There are some interesting PDFs there. Chew, mull, and consider.

Also consider what Horst Simon, head of NERSC said here too.

Never mind Teraflops, we should have a measure of web server load called "Slashdots".

Today SGI announced the Origin 3900 server, the world's densest computer. How dense? How about 16 MIPS R14000A processors and 32 GB of RAM in a 4-rack-unit 'superbrick,'.

The server may be SGI's densest, but at least as far as processing power, it is not the densest. As a counterexample, the above configuration has four processors per unit. Many vendors sell 1U Athlon servers in which each unit holds two dual Athlon systems (four processors per unit), and I can assure you that an AthlonMP 2200+ is quite a bit faster than a MIPS R14000 @ 600MHz.
True, those two Athlon systems aren't a single server, but we're talking density here.
Regardless, SGI does have the Athlon beaten hands down on memory per unit.

I believe that IBM using their Blue Gene to do computational gene folding simulations. It appears that they have a vested interest in biology. Some older articles explains IBM's initiative.

Not theoretical, read here.
here.
and here.
as well as many others. Just because there aren't currently widely deployed applications for things like these, doesn't mean that there won't be in the future. It makes sense to prepare for such situations now.

And just so I can be a troll, you need to capitalize 'Internet'.

I imagine that any self respecting country would have some kind of indigenous dino maker. Let's see. Germany? Nope. UK? Nope. Similar pages can be found for France. Bully for Germany and Japan for at least trying, but it looks like the US kicks ass in this field. I suppose that you can charge alot when you make something others have a hard time keeping up with.

We shall see the merits of the case.

"Washington 19 Oct 99 Republican presidential candidate Gary Bauer called on Gov. George W. Bush to reverse his position calling for an ease on supercomputer export controls."

Googling around, I see a lot of right-wing wackiness attacking both Bush Jr. and Clinton for proposing (and actually doing, respectively) the lifting of supercomputer restrictions. One 1999 report called "RED FLAGS OF TREASON" suggests that China is pretending to know more about supercomputers than it really does, so that the gullible Americans will let down their guard and sell them the supercomputers they can't make themselves.

Now that Apple sells "supercomputer" laptops and Cringely is writing about building a clustered supercomputer in his garage, the restrictions of the 80s and 90s seem a little silly...

As much as we cajole the Australians for their backwards-looking regulations from time to time, you have to hand it to them: we in the US would've never thought of this innovation in government. (Who can argue that government is no longer relevent with such pioneering efforts in the world around us?) But I (like many of my colleagues) fear that the US (beholden to our written constitution) may be too slow to close the gap in this new "Toilet Race".

In order to speed the process of toilet locating, every citizen must be equipped with a swallowable camera. Once all the cameras have been deposited, it will be a simple matter of "looking around" and the project will complete itself. Forty years ago, JFK sacrificed his life to ensure that we would not lose the space race to the Russians. We must not be any less dilligent in our own time.

The Japanese are carrying out an insanely ambitious project,for a 640 node, 40 sustained TeraFlops computer, housed in a building the size of a large hockey arena. They call it the "Earth Simulator" and its main purpose is to carry out atmospheric/climatological research and simulations of the simmering ball of lava we live on (volcano and earthquake research).

Construction is in full swing now; hardware to come online first quarter 2001, software "will take a little longer".

More tech-oriented info here.

Maya seems to cost more but it also does more out of the box. (plus is easier to use than Max) Once you add in enough modules to Max to bring it to the same level of capability, the cost ends up being fairly similar. Max is a great package but if you are going to compare the two, make it an apples to apples comparison.

Check here for more details.