Design World

  • Home
  • Technologies
    • 3D CAD
    • Electronics • electrical
    • Fastening & Joining
    • Factory automation
    • Linear Motion
    • Motion Control
    • Test & Measurement
    • Sensors
    • Fluid power
  • Learn
    • Ebooks / Tech Tips
    • Engineering Week
    • Future of Design Engineering
    • MC² Motion Control Classrooms
    • Podcasts
    • Videos
    • Webinars
  • LEAP AWARDS
  • Leadership
    • 2022 Voting
    • 2021 Winners
  • Design Guide Library
  • Resources
    • 3D Cad Models
      • PARTsolutions
      • TraceParts
    • Digital Issues
      • Design World
      • EE World
    • Women in Engineering
  • Supplier Listings

Google Announces Advanced Chip That Leapfrogs Moore’s Law

By Megan Crouse | May 19, 2016

Share

The Tensor Processing Unit. Image credit: Google

On Wednesday, Google took a look into the future with the opening keynote address at the annual I/O developer conference. They announced several products and services we’ll see rolled out in the next year or so, as well as revealing one which was already in use: Tensor Processing Unit, an ASIC built specifically for machine learning and used to power Google’s datacenters as well as the go-playing AI AlphaGo.

A board with the Tensor Processing Unit can be slotted into a hard disk drive slot on Google’s data center racks and enables them to perform more operations per second, bringing in intelligent results more quickly. Because it’s tailored to machine learning, it is also more tolerant of reduced computational precision, and therefore requires fewer transistors per operation.

“TPU is an example of how fast we turn research into practice — from first tested silicon, the team had them up and running applications at speed in our data centers within 22 days,” Google wrote in a blog post.

The TPU had been secretly in use before its announcement at I/O. It was powering Street View behind the scenes in order to improve the accuracy of the maps, and improved the relevancy of search results with RankBrain.

“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,” Google wrote. They claim that this jumps three generations of Moore’s Law, which states that the number of transistors in a high-end circuit doubles about every two years.

TPUS will also be used to boost developer software like Google’s TensorFlow and Cloud Machine Learning.

At the I/O keynote Google also discussed changes to Android, announced a messaging app called Allo, and a personal assistant and smart home hub similar to Amazon’s Alexa. They’re also bringing a virtual reality platform called Daydream to Android devices, although the headset and controller are still in the preliminary design stages.

They’re also looking for suggestions for the name of the next generation of Android. They’re on top of things and have already said no to “Namey McNameface.”


Filed Under: Rapid prototyping, Virtual reality

 

Related Articles Read More >

PCB mills
Basics of printed circuit board milling machines
Rapid Product Solutions, Inc. enhances its rapid prototyping and production services
Protolabs Launches Production Capabilities for Metal 3D Printing
3D Printer Makes Peacekeeping Missions Cheaper and Repair of Defense Systems Faster

DESIGN GUIDE LIBRARY

“motion

Enews Sign Up

Motion Control Classroom

Design World Digital Edition

cover

Browse the most current issue of Design World and back issues in an easy to use high quality format. Clip, share and download with the leading design engineering magazine today.

EDABoard the Forum for Electronics

Top global problem solving EE forum covering Microcontrollers, DSP, Networking, Analog and Digital Design, RF, Power Electronics, PCB Routing and much more

EDABoard: Forum for electronics

Sponsored Content

  • Global supply needs drive increased manufacturing footprint development
  • How to Increase Rotational Capacity for a Retaining Ring
  • Cordis high resolution electronic proportional pressure controls
  • WAGO’s custom designed interface wiring system making industrial applications easier
  • 10 Reasons to Specify Valve Manifolds
  • Case study: How a 3D-printed tool saved thousands of hours and dollars

Design World Podcasts

May 17, 2022
Another view on additive and the aerospace industry
See More >
Engineering Exchange

The Engineering Exchange is a global educational networking community for engineers.

Connect, share, and learn today »

Design World
  • Advertising
  • About us
  • Contact
  • Manage your Design World Subscription
  • Subscribe
  • Design World Digital Network
  • Engineering White Papers
  • LEAP AWARDS

Copyright © 2022 WTWH Media LLC. All Rights Reserved. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of WTWH Media
Privacy Policy | Advertising | About Us

Search Design World

  • Home
  • Technologies
    • 3D CAD
    • Electronics • electrical
    • Fastening & Joining
    • Factory automation
    • Linear Motion
    • Motion Control
    • Test & Measurement
    • Sensors
    • Fluid power
  • Learn
    • Ebooks / Tech Tips
    • Engineering Week
    • Future of Design Engineering
    • MC² Motion Control Classrooms
    • Podcasts
    • Videos
    • Webinars
  • LEAP AWARDS
  • Leadership
    • 2022 Voting
    • 2021 Winners
  • Design Guide Library
  • Resources
    • 3D Cad Models
      • PARTsolutions
      • TraceParts
    • Digital Issues
      • Design World
      • EE World
    • Women in Engineering
  • Supplier Listings