How We Scaled Bert To Serve 1+ Billion Daily Requests on CPU

May 26, 2021 03:15 PM (PT)

Download Slides

Roblox is a global online platform bringing millions of people together through play, with over 37 million daily active users and millions of games on the platform. Machine learning is a key part of our ability to scale important services to our massive community. In this talk, we share our journey of scaling our deep learning text classifiers to process 50k+ requests per second at latencies under 20ms. We will share how we were able to not only make BERT fast enough for our users, but also economical enough to run in production at a manageable cost on CPU. Further details can be found in our blog post below:

In this session watch:
Quoc Le, Data Scientist, Roblox
Kip Kaehler, Engineering Manager, Roblox


Quoc Le

Quoc N. Le is a Data Scientist at Roblox, where he is focused on leveraging AI for Natural Language Processing. Prior to Roblox, he designed the machine learning algorithms for world-wide sales foreca...
Read more

Kip Kaehler

Kip Kaehler is an Engineering Manager at Roblox
Read more