How can you properly plan for eg. disk storage requirements etc. Does the tool calculate that upfront via some sort of dry-run mode? For companies with larger datasets this would be a rather important consideration. Also, those backfills will generate a lot of network traffic in clustered environments.