Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
1999-12-30
2003-12-09
Olms, Douglas (Department: 2661)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C370S357000, C370S358000, C370S386000, C370S229000, C370S391000, C370S429000, C710S029000, C710S054000
Reexamination Certificate
active
06661803
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to network devices and particularly to network switches for packet-based communication systems such as Ethernet networks and also to an improved method of operating such a network device or switch. The term ‘switch’ is intended to refer broadly to a device which receives addressed data packets at any of a multiplicity of physical ‘ports’ and which can internally switch those packets so that received packets are forwarded from one or more ports in response to the address data in the packet or modified forms of such address data.
BACKGROUND TO THE INVENTION
Broadly, switches of the kind to which the invention relates have a multiplicity of physical ports at which addressed data packets can be received. The data traffic, and therefore the occupancy of bandwidth, at any particular port may vary widely. Typically, a switch includes means for temporarily storing packets received by the input ports and means, hereinafter called for convenience ‘output buffer’, for the temporary storage of data packets before they are forwarded from a port. Such means might be constituted by individual FIFO stores, by respective allocated memory space within a dynamic random access memory or by other forms of memory. Normally the switch includes a forwarding database which is built up typically by the reading of source addresses in incoming packets. The forwarding database permits, from an examination of a received packet and from recourse to associated data, the determination of the or each port from which a received packet should be dispatched. Packets having destination addresses which do not appear in the database need to be ‘broadcast’ either throughout the whole of the network or part of it in order to achieve resolution of the address. Generally, switches which rely on media access control, otherwise called ‘layer
2
’, addresses are termed ‘bridges’ whereas switches which rely on logical link control, otherwise known as ‘layer
3
’, addresses are normally termed ‘routers’. In almost all switches of this general kind there is a conflict of bandwidths required for different types of packet. In any practical system the buffer space available for the transmitting ports of a switch is limited and accordingly switches in practice are inherently liable to congestion.
Various techniques are currently used to bridge and/or route data packets of different types. These techniques generally rely on the individual packets' protocols or priorities to influence the speed of transmission through the switch and/or to determine whether packets will be discarded if the device becomes congested. Generally, packets of higher priority experience the lower rate of discard.
In British patent application number 9807264.8 filed Apr. 3, 1998 (published as GB-2336076-A) and corresponding U.S. patent application Ser. No. 09/093,287 (Moran et al) there is described a mechanism for limiting the aggregate traffic through a port of a network device. The mechanism essentially comprises a counter which increments in accordance with the sizes of packets through the port and decrements at a rate which determines the maximum allowed data rate through the port. The decrementing rate may be remotely controlled by a network administrator. The traffic through the port is limited, if the content of the counter exceeds a threshold, by initiating ‘flow control’, that is to say the sending of control packets to the other end of the link to which the port is connected. These control packets are, in accordance with a known transmission standard, in a form which instructs the device at the far end of the link to cease sending packets for some period of time specified in the control packet. The proposal does not enable either any distinction to be made between packets of different type or the setting of a maximum bandwidth for traffic of a particular type.
The present invention particularly relates to an improved and more versatile technique for allocating a maximum bandwidth which may be occupied by packets of a particular type in an output link.
SUMMARY OF THE INVENTION
In a preferred form of the invention, bandwidth allocation for at least one defined packet type is achieved using a single buffer memory space. Different packet types are selected by for example snooping the protocol fields of packets and/or priority fields in the packets and categorising the packets according to the selection. For at least one and preferably each selected packet type there is a means which determines whether there are enough credits to store a new packet in the output buffer. Each time a new packet is accepted, a respective aggregate count is decremented by the size of the packet. Meanwhile, the aggregate count is constantly incremented at a selectable or controlled rate. In one embodiment of the invention, a counter may be incremented in response to a proportion of pulses corresponding to or defining the speed of a port. For example, if the desired maximum bandwidth to be occupied by packets of a selected type in an output link is to be 30% of the available bandwidth on that link, a counter for packets of that particular type may be incremented in response to three out of each ten successive clock pulses. Alternatively pulses corresponding to slot pulses for an internal slotted bus may be used if it is desired that the counting be independent of the speed of the port.
In an idle state, the counter may be incremented up to the allocation of the output buffer and may thereby store enough ‘credits’ for bursty traffic.
REFERENCES:
patent: 5604867 (1997-02-01), Harwood
patent: 6094435 (2000-07-01), Hoffman et al.
patent: 6098110 (2000-08-01), Witkowski et al.
patent: 6108306 (2000-08-01), Kalkunte et al.
patent: 6122251 (2000-09-01), Shinohara
patent: 6285658 (2001-09-01), Packer
patent: 6347091 (2002-02-01), Wallentin et al.
patent: 6412000 (2002-06-01), Riddle et al.
patent: 6418118 (2002-07-01), Hay et al.
patent: 6434137 (2002-08-01), Anderson et al.
patent: 6456590 (2002-09-01), Ren et al.
patent: 6493315 (2002-12-01), Simpson et al.
patent: 6496478 (2002-12-01), Choi et al.
patent: 0753950 (1997-01-01), None
patent: 2280574 (1995-02-01), None
patent: 2336076 (1999-10-01), None
patent: WO98/09408 (1998-03-01), None
Allwright Gareth E
Choi Kam
Gibson Patrick
Hay Christopher
3Com Corporation
Nguyen Van
Nixon & Vanderhye P.C.
Olms Douglas
LandOfFree
Network switch including bandwidth controller does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Network switch including bandwidth controller, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Network switch including bandwidth controller will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3136682