Import from S3 (External)

    Reference data files in S3 as an external table. No data is copied.

    External Table

    S3 imports create external tables. They do not include native version control, but they can be refreshed via a Glue crawler schedule.

    Step 1: Provide the S3 Location

    Enter the S3 path (for example, s3://my-bucket/path/to/data/). Choose the file format and partition style:

    • Format: Auto, Parquet, CSV, JSON, Avro, ORC
    • Partitions: Auto, Hive, or None
    • CSV Options: Delimiter and Header (shown when CSV is selected)
    Screenshot placeholder: S3 location input, format/partition controls, detect button
    Configure the S3 path and parsing options before detection.

    Step 2: Detect Schema and Preview

    Click Detect to validate the path and infer schema. Review the preview table and confirm the detected columns and partition fields.

    Step 3: Permissions and Schedule

    Apply the generated S3 bucket policy so Nile can read the data. Set a crawler schedule to monitor schema changes.

    Step 4: Create the External Table

    Provide a table name and description, then click Create External Table.