Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
1999-10-27
2004-06-08
Chin, Wellington (Department: 2664)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
Reexamination Certificate
active
06747979
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to a method and apparatus for bridging between networks. More particularly, the invention relates to a method and apparatus for connecting computer networks of dissimilar types.
2. Description of the Related Art
This text refers to IEEE standards, Requests for Comments (RFCs) and Internet-Drafts. These are all standard sources in the networking field. IEEE standards are published by the Institute of Electrical and Electronics Engineers, Inc., 345 East 47
th
Street, New York, N.Y. 10017, USA. RFCs are a series of notes about the Internet, and include the specification documents of protocols associated with the Internet as these are defined by the Internet Engineering Task Force (IETF). RFCs are published by the Information Sciences Institute (ISI) of the University of Southern California (USC). Internet-Drafts are working documents of the Internet Engineering Task Force (IETE) and its working groups (and of certain other bodies) and may form the basis for later RFCs. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. Both RFCs and current Internet-Drafts are obtainable through the World Wide Web site of the IETF, http://www.ietf org/.
Computer networking is complex, and for practical tractability is divided into a number of subtasks. Conventionally, a network is divided into layers, with each layer being responsible for providing a service to the layer above it, the layer itself calling on the services of the layer below it.
The generally accepted model for networking is the OSI (Open Systems Interconnection) Reference Model defined by ISO. This defines seven layers, as are illustrated in FIG.
1
. Each layer
1
at a node
4
communicates with its peer layer
1
at another node through the use of a protocol
2
. Such communication is accomplished through direct communication with the layer below. The communication
3
between overlying layers is known as an interface.
The layers as defined in the OSI are as follows.
Physical Layer
This operates to transmit unstructured bits of information across a link. It is relevant to similar fundamental structural arrangements such as connector type and identification of the purpose of different wires in a cable.
Data Link Layer
This operates to transmit basic structural blocks of information across a link. This is the critical layer for communication within a local area network (LAN), and deals with addressing within a LAN, for example. A sublayer of the data link layer is the medium access control (MAC) layer, which addresses issues specific to a particular type of LAN.
Network Layer
This operates so as to enable any pair of systems in the network to communicate with each other. The dominant network layer protocol is the Internet Protocol (IP). The network layer is responsible for issues such as route calculation and packet fragmentation and reassembly.
Transport Layer
This operates to achieve a reliable communication stream between two systems. The transport layer deals with issues such as lost packets and packet reordering.
Session Layer, Presentation Layer, Application Layer
These are used for higher level services (particular communication patterns, data representations, standardisation of applications) and are not relevant to the communication issues under consideration here.
Communication within a single LAN is handled by the data link layer. A basic problem in networking is communication between two or more LANs, or between a LAN and a network backbone. If two LANs to be connected together are of the same type or sufficiently similar and share the same MAC level addressing, then a bridge can be used to link the LANs together. A bridge is a device which connects two LANs (or, rather, nodes on two LANs) at the level of the data link layer. IEEE 802.1d is a standard defining such bridges (termed “transparent” bridges, because nodes on the network are unaware of the existence of such bridges, the nodes “seeing” other nodes directly across the bridge)—bridges not conforming to the IEEE 802.1d can of course be constructed, but are not standard network components. The operation of a basic bridge is described below with reference to FIG.
2
.
The bridge
21
connects two LAN segments
22
,
23
, accessing each segment via a separate port
25
,
26
. Each LAN segment has a number of nodes
24
. The bridge
21
listens to every packet transmitted on either of the LAN segments
22
and
23
. For each packet received, the bridge stores the MAC address in the packet's source address field in a cache, together with the port on which the packet is received. The bridge
21
then looks through its cache to find the MAC address in the packet's destination address field. If the destination address is not found in the cache, the packet is forwarded out through all the ports except the one on which it was received. If the destination address is found in the cache, the packet is forwarded only through the appropriate port—however, if the “appropriate port” is the one on which the packet was received (meaning that the packet was for transmission between nodes on a single LAN segment), the packet is dropped.
The effect of this functionality is that the bridge can learn MAC addresses, and does not require configuration. For example, consider that bridge
21
has just been put into place, without knowledge of any node addresses. Say that the first packet sent is from node A to node B. This packet will be received by the bridge
21
through the port
25
. The bridge
21
will then store in its cache the MAC address of node A, which is in the source address field of the packet, together with the datum that this address is accessible through the port leading to LAN segment
22
. The packet will then be forwarded on all other ports (in this case, the port
26
leading to LAN segment
23
), as the bridge will have no record of the address of node B in its cache—in this case, this communication is unnecessary for receipt by node B, as node B is on the same LAN, segment as node A. Say that the next packet is sent by node D to node A. The packet will be received by the bridge
21
through the port
26
corresponding to LAN segment
23
, and the MAC address of node D and the datum that this address is accessible through the port
26
leading to LAN segment
23
are recorded in the bridge cache. The destination address, that of node A, is already in the bridge cache. The bridge
21
will thus transmit the packet out through the port
25
connecting to LAN segment
22
, and would not transmit it through any other port. If the next packet is from node B to node A, the bridge
21
will capture the relevant data for node B in the cache—it will also know from the cache that node B and node A are on the same LAN segment
22
, and hence will not forward the packet at all.
The bridge has considerable advantages: it requires (in its basic operation) no configuration, as it can operate immediately and learn the information it requires to operate with full efficiency, and it is transparent to the nodes (in that the node does not perceive any difference between communication on its own LAN segment and communication through the bridge to another LAN segment). Such a bridge does however rely for its operation on similar MAC level addressing on each LAN segment, as otherwise messages forwarded across the bridge will be unintelligible. Many types of bridge design have been proposed, though alternative designs differing from that described above do not fall within the IEEE 802.1d standard. One particular design was proposed in RFC 925 by J. Postel of ISI. The RFC 925 scheme proposes bridging between LAN segments, but storing Internet (IP) addresses, rather than simply MAC level addresses, in the bridge and using network layer protocols to effect bridging. A conventional bridge is a more effective choice than an RFC 925 bridge for connecting LAN segments of similar type locally because use of a network pr
Banks David
Smith Duncan
Wiley Anthony John
Chin Wellington
Hewlett-Packard Development Company L.C.
Schultz William
LandOfFree
Method and apparatus for bridging between networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for bridging between networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for bridging between networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3360781