Google Makes Its Special A.I. Chips

Feb 13, 2018 | 3057 Views

Google created a new kind of computer chip to help power its giant artificial intelligence (AI) systems. These chips were designed to handle the complex processes is believed to be a key to the future of the computer industry.

Google would allow other companies to buy access to those chips through its cloud-computing service. Google hopes to build a new business around the chips, called tensor processing units, or T.P.U.s.
"We are trying to reach as many people as we can as quickly as we can," said Zak Stone, member of designing team of Google engineers. 

Google's move highlights several sweeping changes in the way modern technology is built and operated. It is the initiator for designing chips specifically for AI, providing worldwide platform for a dozen of start-ups as well as familiar names like Intel, Qualcomm and Nvidia.

Google designs much of the hardware inside the massive facilities, from the computer servers to the networking gear that ties the machines together. This reduces cost and improves the efficiency of the multibillion-dollar data centers that supports its online presence.

In addition to its T.P.U. chips, it has also designed an A.I. chip for its smartphones.

Google's new service is focused on a way to teach computers to recognize objects, called computer vision technology. But as time goes on, the new chips will also help businesses build a wider range of services, Mr. Stone said.

Lyft, a driverless car project began testing Google's new chips hoping to accelerate its work on their product. Using the chips, it wanted to accelerate the development of systems that allow driverless cars to, say, identify street signs or pedestrians. Training these systems can take days, and with the help of new chips, it can also be reduced to hours.

Google's T.P.U. chips are housed inside its data centers. The company designs much of the hardware inside the multibillion-dollar data centers as a way of cutting costs and improving efficiency. It has helped to accelerate the development of everything, right from the Google Assistant, to Google Translate, the internet app that translates one language into another.

It is also reducing Google's dependence on chip makers like Nvidia and Intel. In a similar move, it designed its own servers and networking hardware, reducing its dependence on hardware makers like Dell, HP and Cisco.

"This keeps costs down, which is essential when running a large online operation", said Casey Bisson, who helps oversee a cloud computing service called Joyent, owned by Samsung. 

A new wave of artificial intelligence, including services like Google Assistant, are driven by ??neural networks'. They are complex algorithms that can learn tasks on their own by analyzing vast amounts of data. 

Typically, engineers train these algorithms using graphics processing units, or G.P.U.s, which are chips that were originally designed for rendering images for games and other graphics-heavy software. Most of these chips are supplied by Nvidia.

In designing its own A.I. chips, Google was looking to exceed what was possible with these graphics-oriented chips, speed up its own A.I. work and lure more businesses onto its cloud services.

At the same time, Google has gained some independence from Nvidia and an ability to negotiate lower prices with its chip suppliers.

"Google has become so big, it makes sense to invest in chips," said Fred Weber, who spent a decade as the CTO at the chip maker AMD. "That gives them leverage. They can cut out the middleman."

This does not mean that Google will stop buying chips from Nvidia and other chip makers. But it is altering the market. "Who's buying and who's selling has changed," Mr. Weber said.

Mr. Weber and other insiders question whether Google would ever do this, just because a C.P.U. is so complex and it would be so much more difficult to design and maintain one of these chips. But at a private event in San Francisco last fall, David Patterson, a computer science professor at the University of California, Berkeley, who now works on chip technologies at Google, was asked if the company would go that far.

"That's not rocket science," he replied.

Source: HOB