As the owner of a large repository, I need to create (as a sequence of calls to the REST API) 10M objects which have a total disk footprint of size S (e.g. 1TB). On a single node Fedora, this requires time T (e.g. 10 weeks). I would like to be able to add as many nodes as necessary to achieve ingest time T/10 (e.g. 1 week).