Cisco Technology, Inc.

Dean Cheng

Snehal G Karia

Kenneth Durazzo

Paul M Quinn

Andrew George Harvey

Robert E Gleichauf

Richard Mandfred Pruss

Frederick Juergens Baker

Satish K Gannu

Saravanakumar Rajendran

Sandeep Dhar

ShuHung Alexander Mak

Barry Raveendran Greene

In one embodiment, an apparatus can include a service broker configured to: (i) register a service classifier, and to provide context information to the service classifier; and (ii) register a plurality of service nodes. The service broker can also receive capability and service requests from the service classifier. Further, the context information can include a service header, a reachability indication, and an encapsulation, where the service header and the encapsulation may be attached or related to a packet in the service classifier. In addition, the service classifier can use this information to redirect the packet to a first service node.

Service insertion architecture

A method and apparatus are described for managing congestion in a network. For a receiving node, a congestion status associated with a node in the network is determined. The congestion status is advertised to at least one other node in the network. For a sending node, a congestion status associated with a receiving node in the network is received. The congestion status corresponds to a measured node condition at the receiving node. A call is routed to the receiving node based on the received congestion status.

Managing network congestion using dynamically advertised congestion status

A method, and apparatus to transmit an ATM data message from a source node to a destination node, and reflect at least a portion of the data back t where it is compared to the transmitted data. The destination node includes a reflection circuit service located at a specific selector code address, and the destination node listens for a transmission to the specific selector code for reflecting the portion of received data in a subsequent data message for the connection. The source node addresses a setup message to the destination node containing the specific selector code, and compares the transmitted message with the received message. The source node also includes a method and apparatus to send alternatively trace and path trace information as an attachment to a transmitted message, receive the reflected information element, and pass hop data contained within the information element to a user interface for presentation to a user.

Asynchronous transfer mode system and method to verify a connection

Dynamically created routing table structures which may find application as routing tables for ATM or other computer networks provide a selection of pre-computed routes optimized to a specified constraint such as delay and delay variation (e.g., as measured by CTD and CDV), or administrative weight (AW). The routing tables may be implemented as shortest path trees, e.g., as may be generated by a Dijkstra process, which represent the computed routes within the network, optimized to a specified constraint for one or more specified network service categories. Thus, the routing tables allow for rapid compilation of one or more source-computed routes (e.g., DTLs in an ATM network). In some cases, the routing tables may include alternate routes between network nodes. Any number of tables, between a minimum of three and a maximum of ten, may be created.

Dynamically created service class-based routing tables

A method and apparatus for tracing a virtual circuit. A request message is received from a trace originating node. The request message includes a trace transit list. The trace transit list includes one or more element types in a vendor specific information portion of the trace transit list. At least one of the element types is a node level call processing latency element, a network level call processing latency element, an egress physical port identifier element, an ingress physical port identifier element, an ingress logical port identifier element, or a time stamp element. The one or more element types are added to the vendor specific information portion of the trace transit list. The request message is transmitted toward a trace terminating node. An acknowledge message is received from the trace terminating node. The acknowledge message includes the trace transit list copied from the request message to the acknowledge message by the trace terminating node.

Method and apparatus for tracing a virtual connection

A routing table structure for ATM or other networks comprises a shortest path tree architected and organized as a pseudo-spanning tree. The pseudo-spanning tree is a graphical representation of computed routes within the network, optimized to a specified constraint for a specified network service category, and allows for rapid compilation of one or more source-computed routes (e.g., DTLs in an ATM network) by traversing from a destination node specified in a call request to the calculating node (i.e., the node at which the request is received). Unlike conventional spanning trees, however, the pseudo-spanning tree representation may include alternate routes between network nodes. By providing these alternate routes in a single routing table structure, the present invention may utilize less storage and processing resources as compared to routing table schemes of the past. Further, in addition to the single optimization criterion, the routing table structure includes other network traffic parameters for associated links and nodes of the network.

Routing table structures

A distrubution tree for point-to-multipoint connections within a computer network may be dynamically created according to the number of connections shared by individual point-to-point connections which comprise the point-to-multipoint connection. Each distribution tree represents a collection of DTLs, overlapped from a root node of the tree to leaf nodes thereof. The distribution tree may be organized as a distribution table having two dimensions with row indices thereof comprising PNNI node indices of succeeding nodes along a connection DTL. Individual entries in a first column of the distribution table may indicate PNNI node indices of proceeding nodes along the connection DTL. Individual entries in a second column of the distribution table may indicate the number of point-to-point connections for the point-to-multipoint connection that use a link connecting a pair of proceeding and succeeding nodes. A path (or sub-path) may be selected from the distribution table in response to a request for a point-to-multipoint connection within the ATM network. Further, the distribution table may be updated in response to a call being added or dropped within the network.

Growing community of inventors

Fremont, CA, United States of America

Dean Cheng