The format has changed a bit, since none of our RawShares column is ‘parent’.
But you can test this to be certain.
If your cluster already has jobs pending, have bsmith (who has zero usage) and csmith (who has a lot of usage, relatively) each submit several jobs into
the pending queue. Alternatively, have bsmith and csmith submit jobs with larger resource requests: jobs that are large enough to automatically go into a pending state due to lack of resources. Those might be jobs that request the whole cluster, even.
bsmith’s jobs should get a higher priority as seen from sprio, and bsmith’s jobs should start earlier than csmith’s.