I don't think that's correct. DuckDB already supports concurrent writes within one process. I don't see why this would suddenly serialize all writes.
I don't think that's correct. DuckDB already supports concurrent writes within one process. I don't see why this would suddenly serialize all writes.