Abstract: A method for managing a shared buffer between a data processing system and a network. The method provides a communication interface unit for managing bandwidth of data between the data processing system and an external communicating interface connecting to the network. The method performs, by the communication interface unit, a combined de-queue and head drop operation on at least one data packet queue within a predefined number of clock cycles. The method also performs, by the communication interface unit, an en-queue operation on the at least one data packet queue in parallel with the combined de-queue operation and head drop operation within the predefined number of clock cycles.

Abstract: A method for managing a shared buffer between a data processing system and a network. The method provides a communication interface unit for managing bandwidth of data between the data processing system and an external communicating interface connecting to the network. The method performs, by the communication interface unit, a combined de-queue and head drop operation on at least one data packet queue within a predefined number of clock cycles. The method also performs, by the communication interface unit, an en-queue operation on the at least one data packet queue in parallel with the combined de-queue operation and head drop operation within the predefined number of clock cycles.

Abstract: Systems and methods consistent with the present invention provide a mechanism that can efficiently manage multiple queues and maintain fairness among ports while not placing additional performance demands on the memory used to store the queue data structures. Within a port, high priority traffic is dropped only if it is consuming more than its fair share of bandwidth allocated to that port. Queue arbitration is of low performance cost and simple because it arbitrates only across queues per port, rather than across all the queues in parallel. Accordingly, fair arbitration with relatively little hardware cost.

Abstract: Systems and methods consistent with the present invention provide a mechanism that can efficiently manage multiple queues and maintain fairness among ports while not placing additional performance demands on the memory used to store the queue data structures. Within a port, high priority traffic is dropped only if it is consuming more than its fair share of bandwidth allocated to that port. Queue arbitration is of low performance cost and simple because it arbitrates only across queues per port, rather than across all the queues in parallel. Accordingly, fair arbitration with relatively little hardware cost.