Evolution of ML Fact Store

3 years ago netflixtechblog.com

Summary: This is a summary of an article originally published by Netflix Tech Blog. Read the full original article here →

by https://www.linkedin.com/in/vkaushal21/ At Netflix, we aim to provide recommendations that match our members’ interests.

This post will focus on the large volume of high-quality data stored in Axion — our fact store that is leveraged to compute ML features offline.

Axion fact store has four components — fact logging client, ETL, query client, and data quality infrastructure.

Fig Internal components of AxionAxion’s fact logging client logs facts to the https://netflixtechblog.com/keystone-real-time-stream-processing-platform-a3ee651812a, which outputs data to an https://github.com/Netflix/iceberg table.

DevOps Articles