DataFusion desugars BETWEEN into two comparisons, propagates the implied range through an interval DAG, and estimates selectivity from the ratio of interval cardinalities.
DataFusion rewrites filters into min/max expressions over per-container statistics and evaluates them in one vectorized batch across thousands of files.
DataFusion chooses the join algorithm from predicate shape, not cardinality; statistics then pick the build side, the partition mode, and the output row count.
DataFusion lists object-store prefixes, prunes partitions by path, fetches file metadata concurrently, and packs the surviving files into FileGroups.