## Metadata - Author:: [[robinlinacre.com|Robinlinacre]] - Full Title:: Why Parquet Files Are My Preferred API for Bulk Open Data - Category:: #🗞️Articles - URL:: https://www.robinlinacre.com/parquet_api/ - Finished date:: [[2023-01-10]] ## Highlights - Statically hosted parquet files provide one of the easiest to use and most performant APIs for accessing bulk¹ data ([View Highlight](https://read.readwise.io/read/01gpd3b8r7ye62cfpfqwam9bw5)) - When are static files inappropriate? There are some cases where static parquet files may be inappropriate. Examples include: ([View Highlight](https://read.readwise.io/read/01gpd3h98qttd4a0wv63fyt9et)) - Very large datasets ([View Highlight](https://read.readwise.io/read/01gpd3jzcfjdr8rmtgjdk48qdm)) - **Relational data with a complex schema.** In this case, serving a sqlite or duckdb database as a single file may be a better fit. ([View Highlight](https://read.readwise.io/read/01gpd3kr7r0hfqp11nbbh1hw9f)) - Rapidly changing data to which users need immediate access ([View Highlight](https://read.readwise.io/read/01gpd3m8f0srp4g3wabs1dfsxv))