@User is there a way to set any kind of lifecycle on versioned datasets? I'm not seeing anything in docs about that... Or is Kedro's position that we should use an underlying filesystem's capabilities here (like ZFS snapshots or S3 lifecycle policies)? @User fyi