Monday, February 12th 2007

Intel Research Chip Advances 'Era Of Tera'

SANTA CLARA, Calif., Feb. 11, 2007 - Intel Corporation researchers have developed the world's first programmable processor that delivers supercomputer-like performance from a single, 80-core chip not much larger than the size of a finger nail while using less electricity than most of today's home appliances. This is the result of the company's innovative 'Tera-scale computing' research aimed at delivering Teraflop -- or trillions of calculations per second --performance for future PCs and servers. Technical details of the Teraflop research chip will be presented at the annual Integrated Solid State Circuits Conference (ISSCC) this week in San Francisco.

Tera-scale performance, and the ability to move terabytes of data, will play a pivotal role in future computers with ubiquitous access to the Internet by powering new applications for education and collaboration, as well as enabling the rise of high-definition entertainment on PCs, servers and handheld devices. For example, artificial intelligence, instant video communications, photo-realistic games, multimedia data mining and real-time speech recognition - once deemed as science fiction in Star Trek shows - could become everyday realities.

Intel has no plans to bring this exact chip designed with floating point cores to market. However, the company's Tera-scale research is instrumental in investigating new innovations in individual or specialized processor or core functions, the types of chip-to-chip and chip-to-computer interconnects required to best move data and most importantly, how software will need to be designed to best leverage multiple processor cores. This Teraflop research chip offered specific insights in new silicon design methodologies, high-bandwidth interconnects and energy management approaches.

"Our researchers have achieved a wonderful and key milestone in terms of being able to drive multi-core and parallel computing performance forward," said Justin R. Rattner, Intel's chief technology officer. "It points the way to the near future when Teraflop-capable designs will be commonplace and will reshape what we can all expect from our computers and the Internet at home and in the office."

The first time Teraflop performance was achieved was in 1996, on the ASCI Red Supercomputer built by Intel for the Sandia National Laboratory. That computer took up more than 2000 square feet, was powered by nearly 10,000 Pentium® Pro processors, and consumed over 500 kilowatts of electricity. Intel's research chip achieves this same performance on a multi-core chip that could rest on the tip of a finger.

Also remarkable is that this 80-core research chip achieves a teraflop of performance while consuming only 62 watts - less than many single-core processors today.

The chip features an innovative tile design in which smaller cores are replicated as "tiles," making it easier to design a chip with many cores. With Intel's discovery of new and robust materials to build future transistors and no immediate end in sight for Moore's Law, this lays a path to manufacture multi-core processors with billions of transistors more efficiently in the future.

The Teraflop chip also features a mesh-like "network-on-a-chip" architecture allowing super high bandwidth communications between the cores, and capable of moving Terabits of data per second inside the chip. The research also investigated methods to power cores on and off independently, so only the ones needed to complete a task are used, providing more energy efficiency.

Further Tera-scale research will focus on the addition of 3-D stacked memory to the chip as well as developing more sophisticated research prototypes with many general-purpose Intel® Architecture-based cores. Today, the Intel® Tera-scale Computing Research Program has over 100 projects underway that explore other architectural, software and system design challenges.

Intel is presenting eight other papers at ISSCC, including one which will cover the Intel® CoreTM micro-architecture and its use in dual and quad core processors spanning laptops to desktop PCs and servers, using both 65nm and revolutionary 45nm process technologies. Other papers cover such topics as a Radio Frequency Identification (RFID) reader transceiver chip, a low power cache for mobile applications, a reconfigurable Viterbi accelerator, as well as novel circuits for on-die supply resonance suppression, on-chip phase-noise measurement and adaptive techniques for variations and aging.Source: Intel
Add your own comment

9 Comments on Intel Research Chip Advances 'Era Of Tera'

Well, "here today? OUT OF DATE, LATER TODAY..."


* The "motto of this field"... lol!

(I always felt that patent safes from R&D developments have @ least a 5-10 yr. jump on what things CAN be like, & this is just a look @ that line of thinking!).

Still - not to be a pessimist, but this is NEAT to know, that such power is possible, today, if it were needed, & not in a machine that's the size of your home, as the article tigger69 put out, had put it.


P.S.=> It's part of why I quit buying every-to-every-other year for computer equipment really... that kind of successive buying only keeps you buying, & getting MAYBE 20-30% boosts! I've noted I get around 50% ones IF I held off to buying absolutely NEW/state-of-the-art, every 4-5 years or so. Sometimes, I'll upgrade a vidcard, if the power boost is 'huge' (like 50%, & lately, w/ NVidia @ least (who I pay attention to), it has been that way) but, the prices are usurious/outrageous too... here, I tend to be "a generation behind" until prices drop to where I can handle buying w/ out nuking the bank! apk
Posted on Reply
I buy latest equpiment every 5 years as well. If you're lucky you can double your speed at the rate computing upgrades are coming. Soon phase change won't be so exotic and my next pc probably won't have a single fan in it including the PSU.
Posted on Reply
Completely Bonkers
I like it. I like it.

So is this the direction of intel chipset GPU? Just need to bump up precision from single to double and you've got a 67W GPU >> nVidia 8800 in SLI.

Poor Ageia and their PhysX project. Unless they are able to adapt their APIs to work with a revised Ageia PCI/PCIe card that has this little monster sitting on it. Nice. I want one.


P.S. This is a specialist processor and is not intended to be a "CPU" replcement in the way we understand it on a PC. It's a special purpose device. It can execute hundreds (actually 80) asynchronous calculations as opposed to current parallel vector processors than can handle hundreds of identical computations.

It's an ideal starting point for an ALU (math coprocessor or physics system), or GPU
Posted on Reply

Is the beginning of the end.:confused:
Posted on Reply
FLY3R said:

Is the beginning of the end.:confused:
As said in the title.
It is the Era of Terror:nutkick:
Posted on Reply
Sounds interesting, I wonder what it'll cost. AMD seems to be focusing more on the commercial applications of performance increase; and leaving the Home PC advancement up to Intel. I've been waiting for the new AMD line, but I don't think I want to run a Server, just to get a 4X4. So far, that seems to be all they have. It would be nice to find a L1 mobo with 2X Pci-e 16X, but so far no luck. Now I'm looking at Intel Q. The other problem is that AMD bought ATI and there are no real advancements in ATI chipset mobos for the 4X4 AMD.
Posted on Reply
Easy Rhino
Linux Advocate
but will it let me play crysis??? :nutkick:
Posted on Reply
If we steal it, can we take over the world?
Posted on Reply