DuckDB is not only a storage solution. It can directly query a variety of file formats at rest, without having to re-store anything. That's one of its selling points: you can query across archival/log data stored in S3 (or wherever) without needing to "ingest" anything or double-pay to duplicate the data you've already stored.