Hopsworks provides a S3/ADLS/GCS connector to enable Delta Lake tables to be mounted as an external feature group in Hopsworks. Delta Lake tables can also be a source for feature pipelines, with features stored in Hopsworks’ Online/Offline stores. Delta Lake tables can also be sinks for predictions or logs produced by inference pipelines.
You can extend Databricks with real-time feature serving capability with Hopsworks. You can also centrally manage features stored in a Delta Lake on Hopsworks, enabling them to be joined with features from other data sources. You can use Delta Lake tables as data sources in Python or Spark feature pipelines.