Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
1997-11-07
2001-01-16
Pham, Chi H. (Department: 2731)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C370S463000, C370S466000
Reexamination Certificate
active
06175569
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to data communications networks and more particularly to methods and devices for extending Asynchronous Transfer Mode (ATM) network Quality of Service (QoS) guarantees to Local Area Network (LAN) stations.
BACKGROUND OF THE INVENTION
Networking technologies developed for the purpose of enabling data communications between remote users can generally be characterized as either local area network or wide area network (WAN) technologies. The manner in which data communications are handled in any particular network is quite different depending upon the particular technology employed.
LAN technologies were developed primarily to connect user stations located within a limited geographic area, such as a campus, a single building or even a limited area within a single building. One well known type of LAN technology is Token Ring technology. In a Token Ring network, individual LAN stations are connected to a transmission medium which provides a shared or common ring through which data is transmitted in a single direction. Data inserted onto the ring by a particular LAN station circulates through the ring, eventually returning to the originating station.
Token ring networks transmit data in variable length blocks or frames, each of which has a fixed format header used for routing and frame control. When an originating station wants to send data to another destination station on the ring, the originating station generates a frame including a header with its own address and the address of the destination station. The originating station's own address is included in a Source Address field while the address of the destination station is included in a Destination Address field.
When the originating station gains access to the ring, using a “token” as described below, it inserts the data onto the ring for transmission to the next LAN station on the ring. Each LAN station which receives the frame checks the frame's Destination Address field to determine whether it is the intended destination. If the receiving LAN station is not the intended destination, it forwards the message to the next LAN station on the ring. If the receiving LAN station is the intended destination, it copies the frame into a local buffer while forwarding the message to the next LAN station on the ring. The frame, after circulating around the entire ring, is removed at the originating station.
Token ring networks are considered peer-peer networks, meaning that any station is capable of accessing the ring without first gaining approval of a master station. In Token Ring networks, access control is provided through the use of a token, a special purpose frame, which circulates through the ring. A station acquiring the token is allowed to send a frame but must then send a free token to allow other stations a chance at access to the ring.
While Token Ring networks are quite effective in connecting users in a geographically limited area, there are limits on the number of users that can be connected to a single ring and to the length of that ring. WAN technologies have been developed to enable data communications among large numbers of users distributed throughout large geographic areas, including the entire planet.
Asynchronous Transfer Mode (ATM) is a particular type of WAN technology that has become of considerable interest due to its ability to successfully handle data traffic having considerably different service requirements and properties. A fundamental tenet of ATM technology is that all data, no matter what it represents, is transported in fixed length data units, commonly referred to as cells. The mandatory use of fixed length cells makes it possible to switch data at high speeds, using special hardware-based switches forming switching points distributed throughout the network.
Although the ATM cell structure may be standardized, the data transported in the cells can represent different types of information having different, sometimes inconsistent characteristics and transport requirements. Data being switched through an ATM network may represent alphanumeric information, audio (such as voice or music) information, and/or video (such as “live” video, photographs, graphic images or scan-producing medical images) information.
All types of data can be characterized in terms of tolerance to data losses and delays during transmission. A given type of data is considered loss-tolerant if moderate losses during transmission do not significantly degrade the information reconstructed from the data at its destination. Audio information is probably the best example of loss-tolerant information. Similarly, a given type of data can be characterized as delay-tolerant if moderate delays during transmission do not impair the usefulness of the information at the destination. Alphanumeric information is generally a good example of delay-tolerant information. Data having different loss and delay tolerances must be handled or serviced differently during transmission to avoid degradation of information reconstructed from the data at its destination.
To accommodate the different service requirements of different kinds of data traffic which can coexist in an ATM network, different classes of ATM traffic have been defined. The defined classes of interest are CBR (Constant Bit Rate), VBR (Variable Bit Rate), ABR (Available Bit Rate) and UBR (Unspecified Bit Rate).
CBR traffic is traffic that needs to be transmitted at a predictable, substantially constant bit rate. Examples of CBR traffic include real-time audio or video traffic or an ATM network connection emulating a standard Ti (1.544-Mbit/s) line. CBR traffic is delay and jitter intolerant. During setup of a CBR connection, a Peak Cell Rate (PCR) must be negotiated to define the maximum rate at which data can be transported without creating a significant risk of cell loss. Data is typically then transmitted at the negotiated PCR rate. If an attempt in made to transmit traffic at a rate exceeding the PCR, the excess traffic may be discarded by the network.
VBR traffic includes two subclasses of traffic - VBR real-time (VBR-RT) and VBR non-real-time (VBR-NRT). VBR-RT traffic is traffic which may be generated at varying rates (that is, be bursty) while still being subject to tight limits on acceptable cell jitter; that is, cell-to-cell variations on arrival times. Examples of VBR-RT traffic include video signals generated by a variable-rate codec or aggregated voice traffic with silence removal. VBR-NRT traffic is traffic which may be bursty but which is more delay tolerant than VBR-RT traffic. An example of VBR-NRT traffic includes traffic resulting from transaction processing, such as credit verification or other point-of-sale operations.
For each VBR connection, a Peak Cell Rate, a Sustained Cell Rate (SCR) and a jitter tolerance value is negotiated during the connection setup process. The negotiated SCR represents an upper bound for the average throughput over the connection. While traffic can be accepted from a VBR source at rates exceeding the negotiated SCR for short periods of time (as long as the excess rates don't exceed the negotiated PCR), the rate at which traffic will subsequently be accepted from the same source must be reduced below the SCR sufficiently to maintain the negotiated SCR over a longer period of time. To assure that the SCR parameter can be observed over a relatively long period of time, still another parameter, a burst tolerance, is established when the connection is being set up. Burst tolerance defines how long a VBR connection will be allowed to accept traffic at rates greater than SCR before the traffic rate is reduced below SCR to maintain an overall throughput not exceeding SCR.
ABR service attempts to exploit the availability of network bandwidth that becomes available due to the lack of CBR or VBR traffic. ABR implementations utilize traffic management techniques to monitor actual or incipient network congestion which might, if not taken into account, lead to unacceptable cell loss during tr
Ellington, Jr. William W.
Noel, Jr. Francis E.
Tomek Lorrie A.
Duong Frank
International Business Machines - Corporation
McConnell Daniel E.
Pham Chi H.
LandOfFree
Extending asynchronous transfer mode ATM QoS across local... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Extending asynchronous transfer mode ATM QoS across local..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Extending asynchronous transfer mode ATM QoS across local... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2556338