Databricks Feature Store now delivers real-time features with sub-second freshness
Databricks Feature Store now provides sub-second freshness for ML features, reducing feature lag from minutes or hours to 200ms p99 latency. This enhancement enables real-time machine learning use cases such as fraud detection and personalization, which depend on immediate insights. Data scientists and ML engineers can leverage fresh signals without building complex custom streaming infrastructure. The capability is powered by Spark Real-Time Mode for continuous processing and Lakebase for streaming-optimized online storage.
Features (1) ›
- Sub-second Feature Freshness for Real-time ML
The Databricks Feature Store now delivers streaming aggregations from Kafka to the online store with 200ms p99 latency, collapsing feature lag from minutes or hours to milliseconds. This enables critical real-time ML use cases like fraud detection and personalization. The capability is built upon Spark Real-Time Mode (RTM) for continuous processing and Lakebase for streaming-optimized online storage.
https://www.databricks.com/blog/how-databricks-feature-store-serves-features-sub-second-freshness
Related releases
- Databricks Simplifies SQL ETL with Declarative Patterns in Lakehouse Databricks Blog ·
- Databricks Details AI SRE Platform for Accelerated Incident Investigation Databricks Blog ·
- Run, Debug, and Scale Databricks Workloads from Local IDEs Databricks Blog ·
- Databricks Terraform Provider v1.129.0 Enhances Resource Export Capabilities Terraform Databricks Provider Releases ·
- Databricks Java SDK v0.150.0 Enhances Unit Test Fixture Portability Databricks Java SDK Releases ·
- Databricks Runtime 13.3 has reached end of life endoflife.date ·