Day 1
Oct 12, 2021
1:40 pm

Databricks Feature Store Co-designed with a Data and MLOps Platform

The team at Databricks talk about the motivations and use cases of feature stores across different industries.

About this session

Databricks Feature Store is the first of its kind that has been co-designed with Delta Lake and MLflow to accelerate ML deployments. It inherits all of the benefits from Delta Lake, most importantly: features stored in Delta Lake in an open format, built-in versioning, and automated lineage tracking to facilitate feature discovery. By packaging up feature information with the MLflow model format, it provides lineage information from features to models, which facilitates end-to-end governance and model retraining when data changes. At model deployment, the models look up features from the Feature Store directly, significantly simplifying the process of deploying new models, features, and user applications.

In this talk, we will discuss motivations and customer use cases from a variety of industries and the core principles behind the design of the Databricks Feature Store. Furthermore, we will demonstrate how the platform enables our customers to discover and reuse features with automatic lineage tracking, customize feature computation, use open formats for storage and lookup of batch and online features, and eliminate online/offline skew with native model packaging.

Add to Calendar

Join our Slack channel to stay up to date on all the latest feature store news, including early notification when the conference details emerge.