# Sample Avro file for testing schema-aware pipelines, free and CC0 1 Avro test file of 109.2 KB. A direct URL, the file's SHA-256 and CC0, so curl and CI can fetch it with no signup and no keys. The page: . Apache Avro is a row-oriented format common in Kafka pipelines and Hadoop jobs. Every Avro data file embeds the writer's schema in its header, so a reader can decode the records without a separate schema registry. `avro/people-1000.avro` holds 1,000 rows of the shared people dataset, deflate-compressed. Because the schema travels with the data, this file is useful for testing schema resolution: read it with a different reader schema and check how added, removed or renamed fields are handled. Because the rows are the same records as the CSV, JSON, Parquet and Arrow people files, you can also check that a pipeline produces the same result whichever format it reads. Deflate is one of the codecs the Avro specification requires every implementation to support, so the file should open in any conforming reader. Related formats: [Parquet](https://loremfile.dev/parquet) for columnar storage and [Arrow](https://loremfile.dev/arrow) for in-memory interchange. ## Files | Path | Size | MIME | Properties | |---|---|---|---| | [`avro/people-1000.avro`](https://loremfile.dev/avro/people-1000.avro) | 109.2 KB (109,157 bytes) | `application/avro` | 1,000 rows, 14 columns | The same list as JSON: .