Processor manufacturers need to focus on adding features to their products, an IBM leader maintains.
The days of relying on shrinking transistors to achieve performance gains are over, and the chip industry needs to enter a new era of innovation where system-level features are just as important as thinner transistor gates, said IBM's head of processor research, Bernie Meyerson, in a keynote address at the Fall Processor Forum.
For years, chip designers have been enabling huge increases in processor performance by sticking very closely to a two-year cycle of process technology reductions, said Meyerson, (who is vice president and chief technologist at IBM's Systems and Technology Group). Smaller transistors allowed chip designers to crank up the clock speeds, add more cache memory and reduce the size of their processors without having to change many features from generation-to-generation, he said.
Power leakage gets critical
The advent of 90 nanometre process generation has changed that strategy for most chip makers, Meyerson said. Chips are now so small that atom-level defects on a silicon chip can cause power leakage up to 100 times the normal level, he said.
While current designs rely on innovations within the processor, future performance increases will be about chip and system-level innovation, Meyerson said. These future innovations include dual processing cores, embedded memory and software, he said.
"This industry has begun to travel down a road not travelled. We need to identify the things that will win in the future," Meyerson said.
Problems with power leakage at 90 nm and industry leader Intel's shift away from skyrocketing clock speeds have been top concerns for every major microprocessor vendor and analyst firm over the last year.
Move to Plan B
In the 1980s, CMOS (complementary metal oxide semiconductor) transistors replaced bipolar transistors in order to keep the rate of processor innovation on track, Meyerson said. The industry needs a similar type of "Plan B" right now, but there are many different opinions as to how that shift should be accomplished, he said.
Dual-core designs are one way that the industry hopes to keep performance on track. IBM has had a dual-core processor since the introduction of the Power 4 in 2001, and much of the industry is planning to follow in those footsteps, Meyerson said. Two individual processor cores running more slowly than a single-core processor can outperform that chip without a huge increase in power consumption.
However, chip makers must avoid the temptations to cram as many processor cores as possible onto a chip, Meyerson said. This would eventually lead the industry down the same path blazed by faster and faster single-core designs, where relying on a single method eventually runs into a wall, he said.
New tech looks good in Blue Gene
As might be expected, Meyerson pointed to an IBM product as an example of how processors need to evolve. IBM's recent Blue Gene supercomputer is powered by thousands of chips that run much slower than many processors, but take advantage of system-level engineering to enable performance, he said.
The Blue Gene supercomputer built for Lawrence Livermore National Laboratories recently posted a Linpack score of 36.01 teraflops, or 36 trillion floating point operations per second, enough to reclaim the title of the world's faster supercomputer. But the most interesting story about Blue Gene is that it achieved that performance with a system that was 1/100th the size of the Earth Simulator machine that it edged out, Meyerson said. It also consumed 1/28th as much power as the Earth Simulator, he said.
Transistors will continue to shrink, Meyerson said. IBM, Intel and other chip companies are steadfast in their determination to shrink process technology generations every two to three years. But recent chip-making innovations such as strained silicon and silicon-on-insulator technology will grow more important with each successive process generation shrink and technologies such as virtualization will become widespread, he said.
"There are trajectories forward that are enormously promising," Meyerson said.