Skip to main content
Many object storage platforms offer “S3 compatibility” enabling writing to and reading from the object storage using the S3 protocol. The S3 protocol uses an HMAC key comprised of an access ID and a secret to authenticate and write data.

Step 1: Create an HMAC Access ID and Secret

Consult your object storage platform’s documentation to learn how to generate an HMAC Access ID and Secret.

Step 2: Add your destination

Securely share your bucket host, bucket name, chosen folder name, HMAC access ID, and HMAC secret with us to complete the connection.

Permissions checklist

  • The HMAC access ID and secret have write permissions to the target bucket.
  • The bucket is accessible from the data-syncing service’s egress IP.

FAQ

We authenticate using the HMAC access ID and secret you provide. These credentials are used to sign requests to the S3-compatible API. Scope the credentials to only the required bucket to limit access.
Parquet (default and recommended), CSV, JSON, and JSONL.
Files are automatically split; multiple files may be written per model per transfer.
Data lands in Hive-style partitions per model: <folder>/<model_name>/dt=<transfer_date>/<file_part>_<transfer_timestamp>.<ext>. Manifest files are written to a _manifests/ directory at the root of the configured folder path.
Generate new HMAC credentials from your object storage platform and update them in the destination configuration.