Google's New Custom Chip May Not Live Up to the Hype

Google last week announced the Tensor Processing Unit, a custom application-specific integrated circuit, at Google I/O.
Built for machine learning applications, TPU has been running in Google's data centers for more than a year.
Google's AlphaGo software, which thrashed an 18-time international Go champion in a match earlier this year, ran on servers using TPUs.

TPU is tailored for TensorFlow, Google's software library for machine intelligence, which it turned over to the open source community last year.

Moore Still Rules

For machine learning, TPUs provide an order-of-magnitude better-optimized performance per watt, Google said. It's comparable to fast-forwarding technology about seven years -- three generations of Moore's Law.
That claim is misleading, according to Kevin Krewell, a principal analyst at Tirias Research.
"It only works on 8-bit math," he told TechNewsWorld. "It's basically like a Z80 microprocessor in that regard. All that talk about it being three generations ahead refers to processors a year ago, so they're comparing it to 28-nm processors."
Taiwan Semiconductor Manufacturing reportedly has been working on a 10-nanometer FinFET processor for Apple.
"By stripping out most functions and using only necessary math, Google has a chip that acts as though it was a more complex processor from a couple generations ahead," Krewell said.
Moore's law focuses on transistor density and "tends to be tied to parts that are targeted at calculation speed," pointed out Rob Enderle, principal analyst at the Enderle Group. The TPU "is more focused on calculation efficiency, so it likely won't push transistor density. I don't expect it to have any real impact on Moore's law."
Still, the board design "has a really big heat sink, so it's a relatively large processor. If I'm Google and I'm building this custom chip, I'm going to build the biggest one I can put into the power envelope," Krewell noted.
Previous
Next Post »