What is Parquet, and why does DuckDB work well with it?

Parquet is an open, columnar, binary file format designed for efficient storage and retrieval of large analytical datasets, widely used across the modern data ecosystem (Spark, Iceberg, Delta Lake, and many others all read and write it).

Because Parquet is already columnar and stores per-column statistics (min/max values, null counts) alongside compressed data, it aligns naturally with DuckDB's own columnar, vectorized execution engine: DuckDB can use Parquet's embedded statistics to skip reading entire row groups or columns that can't possibly satisfy a query's filter conditions, and can decode Parquet's compressed columns directly into its own internal vectorized format efficiently.

This alignment is a major reason DuckDB has become a popular tool for querying data lakes: pointing DuckDB directly at a folder of Parquet files (locally or in S3-compatible object storage) is often fast enough for exploratory analysis and moderate-scale production queries without needing to first load that data into a dedicated database.