Google Claims Its AI Chips Are Up To 30x Faster Than CPUs And GPUs
During last year's I/O developer conference, Google has revealed the company's own custom chips that will accelerate its machine learning algorithms. The name was revealed, Tensor Processing Units, however, Google didn't go far into explaining what exactly it can do. The only other detail that was revealed is that the TPUs were optimized for the company's own machine learning framework called TensorFlow.
Now, roughly a year after, Google is sharing more data and benchmarks about the said project. Chip designers can find all the specificities of the project in Google's paper. It's worth noting though, that the numbers that the company has declared are all based on their own benchmarking.
According to Google's results, their TPUs are around 15 to 30 times faster in executing their regular machine learning workloads compared to the standard GPU/CPU combo. Since power consumption also matters in a data center, the TPUs are also able to offer 30 to 80 times higher TeraOps/Watt.
Nevertheless, these numbers are all about using machine learning models in production and not about model creation, Tech Crunch reported. Google shares that it began to look into how they could use TPUs in their data center way back a decade ago. However, at the time, there weren't a lot of applications that can benefit from the special hardware yet. Apparently, most of the heavy loads back then were able to settle with using the excess hardware that was already available in the data center.
Google said that things only started changing in 2013 when the company projected that DNNs will become more popular and will ultimately double the demand on the data centers. This would be very expensive to fulfill using the traditional CPUs. This is why the team started the project to produce custom ASIC for inference. The team's goal is to improve cost-performance by up to ten times compared to GPUs.
Google's Cloud Video AI Is Easily Fooled, As Confirmed By Researchers
Google’s new beta technology gets tricked by researchers.
Google Pixel And Pixel XL Automated Bug-fixing Update Fails To Resolve Issues
Google Pixel and Pixel XL continues to have Bluetooth problems despite new Google fix.
Google Leak Points To A Massive Pixel 2017
Google could be working on a massive Pixel device as new leak points to a Google Taimen codename.
HTC Will Build Another Flagship Phone, Even After Billion-Dollar Deal with Google
HTC is showing signs of revival with the announcement of $1.1 billion smartphone business deal with Google.
Google Replacing Defective Nexus 6Ps with Pixel XL for Free
Owners of defective Nexus 6P units in the US and Canada are now receiving Google Pixel XL as free replacement.
MORE IN ITECHPOST
How to Protect Yourself on Facebook
Facebook is an application that takes on an all-encompassing role in our social media lives. It is a space where we communicate with friends and family about our personal lives, browse through an endless stream of content, and interact with other companies and entities on groups and pages.
How Do Personal Emergency Response Systems Work?
Personal emergency response systems, known as PERS for short, are systems that help people to raise the alarm and get immediate help when a medical or personal emergency occurs. They are ideal for older people and anyone with a mobility issue or an injury or illness that can cause falls.
The 9 Most Popular (And Fun) Internet Games of 2019
Are you bored of board games? What are the most popular internet games from this year? These nine are popular for a reason. Check 'em out!