摘要:
At a transmitter, an ATM cell stream is received from the ATM layer and is distributed on a cell-by-cell bases across multiple DSL PHY's. At the receiver, the cells from each DSL PHY are re-combined in the appropriate order to recreate the original ATM cell stream, which is then passed to the ATM layer.
摘要:
A method for flexible inverse multiplexing for ATM and for link grouping of communication links with different transmission rates and delay. The method of the present invention supports links which use a portion of the link bandwidth for one QoS objective (e.g., low delay) and another portion for another QoS objective (e.g., low cell loss). The method involves servicing each of a number of multiplexed flows with a predetermined number of cells. Cells are assigned to each flow during a number of cycles which make up an ATM frame. The number of cells assigned to each flow during each cycle is dynamically adjusted based upon the requirements of each flow, and may be different for each flow. The number of cells assigned to each flow during each cycle may either be fixed, or may vary from cycle to cycle.
摘要:
An advanced SONET add/drop multiplexer (101, 108, 107, 106) capable of supporting packet over SONET and multiprotocol label switching. The add/drop multiplexer (101, 108, 107, 106) is capable of adding and/or dropping both STM and packet traffic, such as Internet Protocol traffic. This SONET add/drop multiplexer allows Internet Protocol streams which are added or dropped at different nodes to be carried over the same SONET path in a network, thereby greatly saving bandwidth.
摘要:
There is provided a fibre channel fabric including (a) a fibre channel interface controller (10) for communicating with a termination node or another fibre channel fabric to control a fibre channel in protocol, (b) an input data buffer (20) for temporarily storing a frame received from the termination node or the other fibre channel fabric, (c) a cell producer (40) for dividing the frame received into fixed length cells, (d) a cell switch (60) for interchanging data at the unit of a fixed length cell, (e) a frame constructor (50) for reconstructing an original frame of the fixed length cells transmitted from the cell switch (60), (f) an output data buffer (30) for temporarily storing a frame transmitted from the frame constructor (50), and (g) a congestion controller (70) for monitoring a load of the cell switch (60) to thereby avoid congestion.
摘要:
A method for flexible inverse multiplexing for ATM and for link grouping (106) of communication links (103, 104, 105) with different transmission rates and delay. The method of the present invention supports links which use a portion of the link bandwidth for one QoS objective (e.g., low delay) and another portion for another QoS objective (e.g., low cell loss). The method involves servicing each of a number of multiplexed flows (300) with a predetermined number of cells. Cells (309) are assigned to each flow (302, 304, 306) during a number of cycles which make up an ATM frame. The number of cells assigned to each flow during each cycle is dynamically adjusted based upon the requirements of each flow, and may be different for each flow. The number of cells assigned to each flow during each cycle may either be fixed, or may vary from cycle to cycle.
摘要:
A system and method are provided for redundancy protection of IMA groups. A protection IMA state machine is synchronized with a working IMA state machine before a hot redundant switch is made. The SCCI number is transmitted directly from the working IMA state machine to the protection state machine over an IMA synchronization link, while delta information is transmitted from the working IMA state machine for use by the protection IMA state machine to derive the frame sequence number.