Decision Mechanics

Insight. Applied.

  • Services
    • Decision analysis
    • Big data analysis
    • Software development
  • Articles
  • Blog
  • Privacy
  • Hire us

Google reveals performance of its custom AI chips

April 6, 2017 By editor

chart showing relative efficiency of different kinds of processing units

Google has just release performance data for its Tensor Processing Unit (TPU) custom machine learning chips.

TPUs are

  • 15-20x faster than GPUs and CPUs when running deep learning applications
  • 30-80x more power efficient (operations per Watt)
  • frugal in terms of lines of code required to control them (their deep learning applications are implemented using 100-1500 lines)

Google claim that without these chips they would have to double the number of data centers they run.

Print Friendly, PDF & Email

Share this:

  • Email
  • Twitter
  • LinkedIn
  • Facebook

Filed Under: Artificial intelligence, Machine learning Tagged With: TensorFlow, TPU

Search

Subscribe to blog via e-mail

Subscribe via RSS

Recent posts

  • Accuracy vs precision
  • It’s not because we have insufficient data…
  • Large Language Models
  • Spreadsheet disasters
  • 10 ways to mislead with data visualization

Copyright © 2023 · Decision Mechanics Limited · info@decisionmechanics.com