Supercomputer On-a-Chip Prototype up and running

[Darksaber]

COLLEGE PARK, Md.-A prototype of what may be the next generation of personal computers has been developed by researchers in the University of Maryland's A. James Clark School of Engineering. Capable of computing speeds 100 times faster than current desktops, the technology is based on parallel processing on a single chip.

Parallel processing is an approach that allows the computer to perform many different tasks simultaneously, a sharp contrast to the serial approach employed by conventional desktop computers. The prototype developed by Uzi Vishkin and his Clark School colleagues uses a circuit board about the size of a license plate on which they have mounted 64 parallel processors. To control those processors, they have developed the crucial parallel computer organization that allows the processors to work together and make programming practical and simple for software developers.

Parallel processing on a massive scale, based on interconnecting numerous chips, has been used for years to create supercomputers. However, its application to desktop systems has been a challenge because of severe programming complexities. The Clark School team found a way to use single chip parallel processing technology to change that.

Vishkin, a professor in the Clark School's electrical and computer engineering department and the university's Institute for Advanced Computer Studies, explains the advantage of parallel processing like this:

"Suppose you hire one person to clean your home, and it takes five hours, or 300 minutes, for the person to perform each task, one after the other," Vishkin said. "That's analogous to the current serial processing method. Now imagine that you have 100 cleaning people who can work on your home at the same time! That's the parallel processing method.

"The 'software' challenge is: Can you manage all the different tasks and workers so that the job is completed in 3 minutes instead of 300?" Vishkin continued. "Our algorithms make that feasible for general-purpose computing tasks for the first time."

Vishkin and his team are now demonstrating their technology, which in future devices could include 1,000 processors on a chip the size of a finger nail, to government and industry groups. To show how easy it is to program, Vishkin is also providing access to the prototype to students at Montgomery Blair High School in Montgomery County, Md.

View at TechPowerUp Main Site

 

[hat]

Finally, something that WON'T bottleneck the 8800

 

[GJSNeptune]

So no more giant heatsinks?

What we really need is to get SSDs big and cheap. Hard drives need a huge boost. Everything else is up-to-speed.

 

[ex_reven]

Finally, something that WON'T bottleneck the 8800

good call

Any idea how far away these are from becoming mainstream?

 

[hat]

I suspect we start seeing them on Newegg within 5 years. By then there will be AGP express and video cards that still use dual-slot coolers, but the developers got smart and made versions that use the small AGP-E slot under it as well as the big AGP-E slot for better performance

 

[GJSNeptune]

Pretty soon video cards will have to be fused with sound cards since they'll be taking up three or four slots for cooling.

 

[d44ve]

So no more giant heatsinks?

What we really need is to get SSDs big and cheap. Hard drives need a huge boost. Everything else is up-to-speed.

Well they just came out with the 64GB SSB

however, it is neither big or cheap... but it is a start

 

[nflesher87]

there aren't even heatsinks on it...or HSFs for that matter...

 

[WarEagleAU]

seems like a move in the right direction

 

[demonbrawn]

Sweet! I wonder who will buy up this technology?

 

[GJSNeptune]

Maybe AMD!

"University of Maryland researchers now heading video card behemoth ATI."

 

[error_f0rce]

Why not use this same technology for video card GPU's too?

The 8800GTSx1000

 

[GJSNeptune]

They kind of do.

 

[newbielives]

So applications don't have to be programmed to support multi threads to take advantage of the 64 parallel processors?

 

[error_f0rce]

They kind of do.

???

Anywho, just stating the obvious I guess. Everything is headed in the multi-core / parallel processing direction anyway.... just hoping to see more development for graphics cards in that area. It seems the focus in the last year or so, has been on multiple cores and heat/power reduction (90nm vs 65nm etc) for CPU's, but haven't seen as much development in those same areas for video cards.

 

[Poisonsnak]

They kind of do.

pics or it didn't happen

 

[GJSNeptune]

That was a really dumb time to use that phrase. Pics of what? Let me go on eBay to find a high-power microscope. Give me a few decades to get the money for it.

A Graphics Processing Unit or GPU (also occasionally called Visual Processing Unit or VPU) is a dedicated graphics rendering device for a personal computer, workstation, or game console. Modern GPUs are very efficient at manipulating and displaying computer graphics, and their highly parallel structure makes them more effective than typical CPUs for a range of complex algorithms.

 

[SK-1]

Yes,..but is it Prime stable???


LOL

 

[Poisonsnak]

That was a really dumb time to use that phrase. Pics of what? Let me go on eBay to find a high-power microscope. Give me a few decades to get the money for it.

Lol and yet you knew exactly what I was looking for (the wikipedia quote). It was a little bit vague but it'll do.

Let me know when you buy that microscope - I might want to borrow it someday. I don't know if you can view transistors in a traditional (optical) microscope though as transistors in a GPU nowadays are in the 80nm range and visible light is more like 500nm. I think you should be able to view it with an electron microscope though.

 

[hat]

I only have 2 transistors.

 

[GJSNeptune]

Lol and yet you knew exactly what I was looking for (the wikipedia quote). It was a little bit vague but it'll do.

Let me know when you buy that microscope - I might want to borrow it someday. I don't know if you can view transistors in a traditional (optical) microscope though as transistors in a GPU nowadays are in the 80nm range and visible light is more like 500nm. I think you should be able to view it with an electron microscope though.

Yes I knew what you wanted, but you didn't get them. And it shows you how much I know about the world at the microscopic level.