Apache DataFusion Blog

Articles by Adrian Garcia Badaracco (Pydantic), Andrew Lamb (InfluxData)

Dynamic Filters: Passing Information Between Operators During Execution for 25x Faster Queries

This blog post introduces the query engine optimization techniques called TopK and dynamic filters. We describe the motivating use case, how these optimizations work, and how we implemented them with the Apache DataFusion community to improve performance by an order of magnitude for some query patterns.

Motivation and Results

The …