The reason our society and economy have become so rapidly and thoroughly digitized in recent years is that the underlying computer processors have been able to keep up, and exceed, all the workloads we’ve thrown at them. It’s Moore’s Law in action. But will there be a point of diminishing growth in processing power?
In 1965, Gordon Moore, co-founder of Intel, postulated that that number of transistors that can be placed on an integrated circuit — and thus processing power — would double every two years.
The rest of the story is the stuff of computer industry legend. It turned out that Moore was right on the money, and the processing power of semiconductors keeps outpacing the workloads we put on them — from high-volume transaction processing to simulations to graphics. At the same time, the form factors that support these applications keep getting smaller and smaller.
But is there a physical limit to how many transistors a minuscule circuit can support? In recent years, however, observers have speculated that we may be near the end of Moore’s Law, at least as it applies to silicon-based processors.
Intel CTO Justin Rattner recently pooh-poohed such speculation, stating in an interview with Network World that Moore’s Law will likely be the rule for many decades to come. “If Moore’s Law is simply a measure of the increase in the number of electronic devices per chip, then Moore’s Law has much more time to go, probably decades,” he is quoted as saying. The future remains in silicon for a long time to come, he adds.
Still, new constructs of processor technology are emerging. For example, IBM announced last month that its scientists, in conjunction with CalTech, developed a method of using DNA molecules as scaffolding — where millions of carbon nanotubes could be deposited and self-assembled into precise patterns by sticking to the DNA molecules. This scaffolding may help enable processors to be etched into surfaces with less than 22 nanometers of space — the current physical barrier of space for cramming transistors on a circuit.
In the silicon world, Intel has already moved from a 45-nanometer construct to a 32-nanometer construct for its processors. The chipmaker also recently unveiled its own plans to crack the 22-nanometer barrier, announcing plans to release the 22-nanometer processors toward the end of 2011.
So, every year, we’ll be packing even more and more power on even tinier chips. Is it just a matter of time before we’re carrying around mainframes within our smartphones?