Since SQLite executing SQL locally on a remote peer machine is essentially computation push-down, one could think of building a planet-scale distributed analyti

Since SQLite executing SQL locally on a remote peer machine is essentially computation push-down, one could think of building a planet-scale distributed analytics engine using such a pattern (perhaps using DuckDB and parquet/arrow files under the hood - but which exact SQL engine is behind the query pushdown API can be abstracted away too) edit: attracted->abstracted