| 摘要 |
A queue management method and system manages queued cells in such a way that higher priority cells are always served first, the low priority cells are dropped when the queue is full and, within the same priority, any interference is prevented. Four different architecture designs for such queue management are presented and their implementation feasibility and hardware complexity are compared and contrasted. A departure sequence is assigned to each cell in the novel architecture to implement the queue management. The sequence applies the concepts of fully distributed and highly parallel processing to schedule cell sending or dropping sequences. Preferably, a sequencer is provided such that the queue size and the number of priority levels can grow flexibly and without limit.
|
