EQM’s fair queuing features
Fair Queuing (round robin) of source port queues within a class of service macro queue.
Weighted Fair Scheduling implementation supports 4 class of service macro queues.
Fairness is applied to classes of service (macro-queues), not individual ports.
Credits are allocated (CBFQ) every service round to accumulators for each class of service macro queue.
Credits are consumed by each packet fragment transmitted.
Bandwidth fairness is a achieved with the granularity of the packet fragment size (240 bytes)
Bandwidth share is determined by the programmable magnitude (weight) of the credit grant to each class of service.
Avoids performing a calculation involving bandwidth share.