Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
2002-07-26
2004-04-06
Kizou, Hassan (Department: 2662)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C709S252000
Reexamination Certificate
active
06717950
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to employing a bridge protocol for interconnecting two or more local area networks (LANs). More particularly the invention relates to apparatus and method, which is backward compatible with existing 802.1D Spanning Tree Bridge Protocol, for improving routing capability of spanning tree forwarding without a significant increase in complexity and which enables traffic prioritized path differentiation and provides load balancing by supporting alternate paths selected according to one of two or more priority values carried by a frame to be forwarded.
2. Background Information
A Local Area Network (LAN) is used to connect end stations together to provide communications. A single LAN permits a limited number of end stations, a limited size, and a limited amount of offered load. In this respect, LANs cannot grow beyond a certain limit. LANs may be interconnected via internetworking devices such as bridges and routers. These devices have different advantages and disadvantages depending on the internetworking environment. In the early days of internetworking, bridges were popular because they were much cheaper and faster than routers. In addition, bridges were used to support heterogeneous network layer protocols. The primitive computing technology of those days favored off-loading of work to larger servers using protocols that were optimized for LANs.
IEEE 802 Standards Committee has specified two bridge protocols. IEEE 802.1 group has issued the IEEE 802.1D Spanning Tree Bridge Protocol and IEEE 802.5 group has issued the Source Routing Bridge Protocol. Among these two schemes, IEEE 802.1D offers a better solution and has been studied more intensively. This approach is transparent to end stations and requires no modifications to the MAC layer of end stations. All the routing related operations are done in the bridges. Today, the IEEE 802.1D Spanning Tree Bridge Protocol is widely used for interconnecting the family of IEEE 802 standard LANs.
A bridge has several ports connecting to different LANs. A frame sent from one LAN to the other will typically go through one or more ports and bridges. As bridges are capable of filtering frames, they are useful for dealing with unnecessary broadcast traffic. Such a broadcast containment capability renders bridging a simple solution to implementing a virtual LAN. This bridged LAN environment should be transparent and looks like a single LAN to end users. The basic function of bridges is to forward MAC (Medium Access Control) frames from one LAN to another, therein providing an extension to the LAN without requiring any modification to the communications software in the end stations attached to the LANs. Bridges do not modify the content or format of the MAC frames they receive. The operation of bridges should not misorder or duplicate frames. Upper-layer protocol transparency is an advantage of bridging since bridges can rapidly forward traffic representing any network-layer protocol without having to examine upper-layer information.
The landscape for internetworking has evolved considerably with advances in high-speed network layer routing and data link layer switching technologies. Functionalities at the two layers are increasingly similar. While routers are generally more intelligent than bridges in terms of their dynamic routing capability, they are also more complicated to operate and costly to implement. Bridges have been designed to span a range of routing capabilities from dynamic source routing to static spanning tree forwarding, thereby allowing a trade-off between routing performance and protocol complexity. Although routers are becoming cheaper and faster than they used to be, they remain more complicated than bridges to operate because intermediate hops must still rise above the data link layer in the protocol stack. In spite of the common wisdom that IP has won the network layer, there are still going to be non-IP network layer protocols in the foreseeable future. On the other hand, while bridges are evolving to accommodate more and more network layer functionality, they will always support multiple network layer protocols.
An IP (Internet Protocol) address encodes both a network and a host on that network. Since it does not specify an individual end system, but a physical location in a network, the IP address of a host must change whenever it moves from one network to another. On the other hand, an IEEE 802 MAC address identifies an end system instead of a physical location in a network, and hence is always applicable to a host no matter where it is located in the network. Such portability of addresses is important particularly for mobility and the benefit of plug-and-play. Although new features, are emerging to minimize the need to configure and reconfigure IP addresses, these features can increase the cost and processing overhead of the system. DHCP (Dynamic Host Configuration Protocol), for example, provides a widely deployed framework for host registration and configuration. DHCP, however, was designed only for fixed hosts on physically secure LANs. DHCP is being extended to allow dynamic reconfiguration of a single host triggered by the DHCP server (e.g. a new IP address). Depending on the bandwidth of the network between server and client, the delay in the reconfiguration process can grow exponentially as failed retransmissions trigger exponential backoff.
The IEEE 802.1D specification defines a protocol architecture for MAC bridges and recommends formats for a globally administered set of MAC station addresses across multiple LANs.
FIG. 1
shows a bridge protocol architecture for a connection of two LANs via local
10
or remote
12
,
14
bridging. Referring to the OSI (open systems interconnect) reference model, a bridge encompasses the first two layers, namely the Physical Layer (layer
1
) and the Data Link Layer (layer
2
). There are two sublayers in layer
2
: Medium Access Control (MAC) sublayer and Logical Link Control (LLC) sublayer. Bridges operate relay functions on the MAC sublayer and interface with the LLC sublayer above through LLC service access points. By using bridges, a growing LAN can be partitioned into self-contained units for administrative or maintenance reasons, as well as to improve performance via load balancing and fault isolation. Bridges are typically used to interconnect LANs of the same type, such as the family of IEEE 802 LANs. Translation among different link-layer protocols is needed, however, when the interconnected LANs are not homogeneous (e.g., IEEE 802.3 and IEEE 802.5 type LANS), and interoperability is achieved by appropriate frame encapsulation.
A bridge relays individual MAC user data frames between separate MAC protocols of the bridged LAN connected to the ports of the bridge. A MAC entity for each port handles all the media access method dependent functions, i.e., MAC protocol and procedures, as specified in the relevant IEEE 802 standard for that MAC technology. Each bridge port receives and transmits frames to and from the LAN to which it is attached using the services provided by the individual MAC entity associated with that port. Each bridge port also functions as an end station providing MAC service to the LLC layer. All MAC entities communicating across a bridged LAN are uniquely identified by their respective 48-bit MAC addresses. A bridge may use a 48-bit MAC address, or a 16-bit locally administered MAC address. This bridge address must be unique within the extended LAN, and a single unique bridge identifier (ID) is derived from it for the operation of a bridge protocol. Each frame transmitted from a source end station to a destination end station carries the MAC addresses of the end stations respectively in the source and destination address fields of the frame's MAC header. A frame that is to be relayed by every bridge to all its neighboring bridges in an extended LAN contains a bridge group MAC address in the destination address field of the frame's MAC header.
Lee Whay Chiou
Lui King-Shan
General Instrument Corporation
Kizou Hassan
Levitan Dmitry
Volpe and Koenig P.C.
LandOfFree
Method and apparatus for priority-based load balancing for... 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 priority-based load balancing for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for priority-based load balancing for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3218062