Intel Teases 80-Core Chip
Intel unveiled more details of its tera-scale computing project on Sunday, saying it hoped to bring Teraflops — or trillions of calculations per second — performance to future PCs and servers. Intel will deliver a research paper on the project this week at the annual Integrated Solid State Circuits Conference (ISSCC) in San Francisco.
Intel first revealed progress of the effort in January. Now it is touting that the performance of the chip will enable futuristic scenarious once only envisioned by the likes of Gene Roddenberry in sci-fi tales.
"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," the company said in a statement.
At the same time, however, Intel tried to temper expectations by saying it doesn't know how far off the future is. Thus, the company has no immediate plan to "bring this exact chip designed with floating point cores to market."
Still, the project has been a learning opportunity for Intel engineers, a chance for them to tinker with the kind of chip-to-chip and chip-to-computer interconnects demanded of such high-performance architectures and experiment with software specialized for multiple processor cores.
Energy management was also a primary interest. The 80-core chip cranked out teraflops of performance while tapping into only 62 watts — less than many single-core processors today, the company said.
"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 Rattner, an Intel senior fellow and chief technology officer. "It points the way to the near future when Teraflops-capable designs will be commonplace and reshape what we can all expect from our computers and the Internet at home and in the office."
Teraflops performance first happened in 1996, on the ASCI Red Supercomputer built by Intel for the Sandia National Laboratory, the company said. Packing in nearly 10,000 Pentium Pro processors, the computer sprawled across more than 2,000 square feet and consumed more than 500 kilowatts of power.
Intel said the next phase of research will focus on putting 3-D stacked memory on the chip, and developing more sophisticated research prototypes using general-purpose cores. As part of its Tera-scale Computing Research Program, Intel also has more than 100 projects underway looking into architectural, software and system design challenges.