AI Cloud & ML Ops Machine Learning & Data Science Research US & Canada

MLPerf Training v0.7 Results Released: Google & NVIDIA Lead the Race

The industry-standard MLPerf benchmark today released the results of the third round of its ongoing ML Training Systems competition.

The industry-standard MLPerf benchmark today released the results of the third round of its ongoing ML Training Systems competition. The competition measures the time it takes to train one of eight ML models to a qualified target on the following tasks: image classification, recommendation, translation, and playing Go. Forerunners Google and NVIDIA set new AI performance records in this third round (v0.7).

image.png
image.png
image.png
image.png

Google TPUs set records in six of the eight benchmarks! We need bigger benchmarks, because we can now train the ResNet-50, BERT, Transformer, & SSD benchmarks each in under 30 seconds, tweeted Google AI Lead Jeff Dean. Google broke performance records in DLRM, Transformer, BERT, SSD, ResNet-50, and Mask R-CNN using its new ML supercomputer and latest Tensor Processing Unit (TPU) chip. Google disclosed in a blog post that the supercomputer includes 4096 TPU v3 chips and hundreds of CPU host machines and delivers over 430 PFLOPs of peak performance.

image.png
Speedup_of_Googles_best_M.0860040916660770.max-2200x2200.png
image.png

While Google ramped up their cloud TPUs and the new supercomputer to deliver faster-than-ever training speeds, NVIDIA’s new A100 Tensor Core GPU also showcased its capabilities with the fastest performance per accelerator on all eight MLPerf benchmarks. NVIDIA introduced the A100 this May, as the first GPU based on the company’s Ampere GPU architecture. With more than 54 billion transistors, the 7-nanometre processor can execute five petaflops of performance.

1-DGX-records-x1280.png
2-A100-records-x1280.png

MLPerf is a consortium of over 70 companies and researchers from leading universities, and the MLPerf benchmark suites are the industry standard for measuring machine learning performance.

For complete results of the MLPerf competition including previous round scores, please visit the project website.


Reporter: Fangyu Cai | Editor: Michael Sarazen


Synced Report | A Survey of China’s Artificial Intelligence Solutions in Response to the COVID-19 Pandemic — 87 Case Studies from 700+ AI Vendors

This report offers a look at how the Chinese government and business owners have leveraged artificial intelligence technologies in the battle against COVID-19. It is also available on Amazon Kindle.

Click here to find more reports from us.


We know you don’t want to miss any story. Subscribe to our popular Synced Global AI Weekly to get weekly AI updates.

0 comments on “MLPerf Training v0.7 Results Released: Google & NVIDIA Lead the Race

Leave a Reply

Your email address will not be published.

%d bloggers like this: