Google's Tensor Processing Unit said to advance Moore's Law seven years into the future

Google unveils a custom chip, which it says advances computing performance by three generations.

Forget the CPU, GPU, and FPGA, Google says its Tensor Processing Unit, or TPU, advances machine learning capability by a factor of three generations.

“TPUs deliver an order of magnitude higher performance per watt than all commercially available GPUs and FPGA,” said Google CEO Sundar Pichai during the company’s I/O developer conference on Wednesday.

TPUs have been a closely guarded secret of Google, but Pichai said the chips powered the AlphaGo computer that beat Lee Sedol, the world champion in the incredibly complicated game called Go.

Pichai didn’t go into details of the Tensor Processing Unit but the company did disclose a little more information in a blog posted the same day as Pichai’s revelation.

“We’ve been running TPUs inside our data centers for more than a year, and have found them to deliver an order of magnitude better-optimized performance per watt for machine learning. This is roughly equivalent to fast-forwarding technology about seven years into the future (three generations of Moore’s Law),” the blog said. “TPU is tailored to machine learning applications, allowing the chip to be more tolerant of reduced computational precision, which means it requires fewer transistors per operation. Because of this, we can squeeze more operations per second into the silicon, use more sophisticated and powerful machine learning models, and apply these models more quickly, so users get more intelligent results more rapidly.”

The tiny TPU can fit into a hard drive slot within the data center rack and has already been powering RankBrain and Street View, the blog said.

Don’t give up on the CPU or GPU just yet

What isn’t known is what exactly the TPU is. SGI had a commercial product called the Tensor Processing Unit in its workstations in the early 2000s that appears to have been a Digital Signal Processor, or DSP. A DSP is a dedicated chip that does a repetitive, simple task extremely quickly and efficiently.

Analyst Patrick Moorhead of Moore Insights & Strategy, who attended the I/O developer conference, said, from what little Google has revealed about the TPU, he doesn’t think the company is about to abandon traditional CPUs and GPUs just yet.

Read more: Moore's Law stutters: Intel officially puts “tick-tock” CPU release cycle on hiatus

“It’s not doing the teaching or learning,” he said of the TPU. “It’s doing the production or playback.”

Moorhead said he believes the TPU could be a form of chip that implements the machine learning algorithms that are crafted using more power hungry GPUs and CPUs.

As to Google’s claim that the TPU’s performance is akin to accelerating Moore’s Law by seven years, he doesn’t doubt it. He sees it as similar to the relationship between a traditional ASIC and a CPU.

ASICs are hard-coded, highly optimized chips that do one thing really well. They can’t be changed like an FPGA but offer huge performance benefits. He likened the comparison to decoding an H.265 video stream with a CPU versus an ASIC for just that task. A CPU without dedicated circuits would consume far more power than the ASIC at that job.

One issue with ASICs though is the cost and their permanent nature, he said. The only way to change the algorithm is if a bug or improvement is found to make a new one. That’s why ASICs have been traditionally the domain of entities with unlimited budgets like governments.

tpu 1

Google said it has been secretly using its TPU to make applications like Street View smarter.

Join the newsletter!

Error: Please check your email address.
Rocket to Success - Your 10 Tips for Smarter ERP System Selection

Tags cpuGoogleMoore's Law

Keep up with the latest tech news, reviews and previews by subscribing to the Good Gear Guide newsletter.

Gordon Mah Ung

PC World (US online)
Show Comments

Cool Tech

SanDisk MicroSDXC™ for Nintendo® Switch™

Learn more >

Breitling Superocean Heritage Chronographe 44

Learn more >

Toys for Boys

Family Friendly

Panasonic 4K UHD Blu-Ray Player and Full HD Recorder with Netflix - UBT1GL-K

Learn more >

Stocking Stuffer

Razer DeathAdder Expert Ergonomic Gaming Mouse

Learn more >

Christmas Gift Guide

Click for more ›

Most Popular Reviews

Latest Articles

Resources

PCW Evaluation Team

Walid Mikhael

Brother QL-820NWB Professional Label Printer

It’s easy to set up, it’s compact and quiet when printing and to top if off, the print quality is excellent. This is hands down the best printer I’ve used for printing labels.

Ben Ramsden

Sharp PN-40TC1 Huddle Board

Brainstorming, innovation, problem solving, and negotiation have all become much more productive and valuable if people can easily collaborate in real time with minimal friction.

Sarah Ieroianni

Brother QL-820NWB Professional Label Printer

The print quality also does not disappoint, it’s clear, bold, doesn’t smudge and the text is perfectly sized.

Ratchada Dunn

Sharp PN-40TC1 Huddle Board

The Huddle Board’s built in program; Sharp Touch Viewing software allows us to easily manipulate and edit our documents (jpegs and PDFs) all at the same time on the dashboard.

George Khoury

Sharp PN-40TC1 Huddle Board

The biggest perks for me would be that it comes with easy to use and comprehensive programs that make the collaboration process a whole lot more intuitive and organic

David Coyle

Brother PocketJet PJ-773 A4 Portable Thermal Printer

I rate the printer as a 5 out of 5 stars as it has been able to fit seamlessly into my busy and mobile lifestyle.

Featured Content

Product Launch Showcase

Latest Jobs

Don’t have an account? Sign up here

Don't have an account? Sign up now

Forgot password?