it optimally allocate jobs for incoming requests to maximize usage of Requests/min and Token/min limits