Slashdot Mirror

"Microsoft is trying to tackle: accessible supercomputing"

Assuming MS was responding to this imagioned problem ..

"The contest showed that supercomputers .. are accessible to people interested in pursuing science, simulation or modeling"

"but the learning curve for getting something running on these machines is pretty intimidating, especially for non-CS based disciplines. I've had to take a 1-2 day class, plus futz around"

You actually programed a supercomouter - cool. What type and where exactly? How does HPC Server differ in respect to other solutions?

"the Blue Gene family of supercomputers has been designed to deliver ultrascale performance within a standard programming environment"

"Hopefully Microsoft can spur the industry in this direction"

You mean like continually inventing Apple, badly .. :)

Read more...

"Wednesday, Apr 11 @ 13:13 PDT The powerful earthquake struck suddenly, shaking the seven-story building so hard it bent, cracked and swayed in response. But this was no ordinary earthquake. In a groundbreaking series of tests, engineering researchers from UC San Diego's Jacobs School of Engineering jarred a full-size 275-ton building erected on a shake table, duplicating ground motions recorded during the January 17, 1994 Northridge earthquake in Los Angeles, California. To record the impact on the building, the structure was fitted with some 600 sensors and filmed as the shake table simulated the earthquake, yielding a flood of data including stress, strain, and acceleration -- so much information that engineers were having a hard time making sense of it all. That's where visualization experts from the San Diego Supercomputer Center (SDSC) at UC San Diego came in. "

http://www.hypertransport.org/products/productdeta il.cfm?RecordID=65
PathScale Infiniband card. Lowest latency infiniband neworking card in existance (1.5 microseconds).

http://www.supercomputingonline.com/article.php?si d=11429
Xilinx card

Articles about HTX and 4x4 (Torrenza) tie in:
http://www.xbitlabs.com/news/cpu/display/200606070 74412.html
http://www.amd.com/us-en/Weblets/0,,7832_8366_5730 ~109409,00.html
http://www.tgdaily.com/2006/06/02/qanda_amd_vp_ran dy_allen/

There are many more, but this is a start.

The University of Buffalo could not run their new Dell cluster at full capacity due to power problems. Power is becoming a HUGE issue in the datacenter.

Your gaming machine at home? Not so much. Hundreds if not thousands of servers in a datacenter (plus all the ethernet and san switches, storage media..etc, etc)...big problems.

A whole deck full of them. Browse through here to get an idea. Uses from nuclear physics, CFD, military applications, applications that will set off the tinfoil beanie crowd...yeah, the .gov still has supercomputers. Heck, how do you think NOAA makes the weather forecasts that AccuWeather wants to (re)sell you?

Sun/TI partnership milestones:
-- 1988 - Sun/TI relationship founded
-- 1992 - Delivery of SuperSPARC(r) and MicroSPARC(r) 1
-- 1994 - Delivery of SuperSPARC II
-- 1995 - Delivery of UltraSPARC I (first 64 bit SPARC processor)
-- 1997 - Delivery of UltraSPARC II (72-way support)
-- 2000 - UltraSPARC III (106-way support)
-- 2001 - Copper UltraSPARC III
-- 2002 - UltraSPARC III (industry's first 64-bit in 130 nm)
-- 2003 - First 64-bit 90 nm process samples

"The choice of InfiniBand as the interconnect for this cluster filled a critical missing piece in creating a world class cluster, based solely upon industry standard computing components," says Hassan Aref, dean of Virginia Tech's College of Engineering. "Unlike its predecessors, InfiniBand is an industry standard and it provides the fastest high performance interconnect available today."

www.supercomputingonline.com/article.php?sid=4316

Not to mention that most machines on the November 2002 list would have faster numbers now, if they re-ran with the Goto BLAS that came out after list enetries were due. Top 80 in the world is probably overstating things a bit.

Goto site is here

Article about the effects is here

That article was a little bit too much opinion, not enough information. This one's a little better:

Just to ease the suspense, he still comes across as a bit of a loony, but at least there is enough meat in the article to properly discuss.

For those of you who are wondering what they mean by high performance networks inside the Q machine..

The Q machine utilizes dual-rail Quadrics card according to this. Dual rail refers to using two NI cards (each one on a separate 64b/66MHz PCI bus so they can get the most out of the I/O system of the host).

I hadn't heard of Quadrics so I looked them up. At the web site you find out that they're a switched network that gets 340 MBytes per second between applications and with latencies around 3-5 microseconds. Compare this to 100Mbps ethernet, which gets 10MBytes/s and latencies of 70+ microseconds and you'll understand why the Q machine will run fine grained parallel apps that the green machine won't be able to touch.

Looking a bit through the literature, I noticed that Quadrics uses IEEE 1596.3 for its link signaling (400 MBaud, 10 bit). While they don't say it anywhere, this IEEE standard is the well-known SCI standard (scalable coherent interconnect.. pretty popular in Europe, but the US has been dominated by Myrinet..which I conicidentally use at school)..

Hope this gives some more detail about the arch..

You can get an SX-6i . The processor in ES is not made only for ES. And I don't think you would sell many supercomputers for IBM if you were advocating Gimp and Mozilla as applications...