Apache Arrow's IPC file format stores columnar data the way Arrow holds it in memory, so a reader can map the file and use it without parsing. arrow/people-1000.arrow contains 1,000 rows of the shared synthetic people dataset, the same records as the CSV, JSON, Parquet and Avro people files, in a single record batch.

Two details matter when testing. First, this is the IPC file format, with a footer, not the streaming format; the two share most of their bytes but are opened by different APIs, and a reader pointed at the wrong one fails. Second, because the rows match the other people fixtures, you can load the same data from several formats and compare the results column by column.

Use it for data pipeline tests, schema inspection, and checks that a service accepting Arrow data handles a file where it might expect a stream. Related formats: Parquet and Avro.

Sample Arrow IPC file for testing data pipelines — free, CC0

Related formats