Ciena Corporation

Omnia Communications, Inc.

Bappa Sinha

Jeffrey N Weiss

Christopher James White

Ming-Teh Ma

Ed Klein

Edward E Klein

James Regan

Walter Dray

William J Regan

Dan Cedrone

A system and a method of operating the system for non-disruptively inserting a node into the operations of an ATM ring uses the ring operations to update topology information and routing tables at the existing nodes to include therein the new node and one or more virtual paths associated with the new node. Basically, the node to be inserted establishes communications with a ring hub node over an established intra-ring management channel. The node and the ring hub node then exchange information over the management channel and the node performs its initializing routines in order that the node may operate to pass through traffic on the existing virtual paths. The new node next requests from the hub node the assignment of one or more virtual paths for directing traffic between that node and other nodes on the network. In response to the request, the hub node assigns one or more virtual paths to the requesting node and establishes a signaling channel to the node over one of the virtual paths. The hub node also notifies the other nodes of the assignment, and downloads to the new node routing tables that include the established connections over the existing virtual paths. The node may then participate in traffic shaping on existing virtual circuits, and also in call set up and call tear down operations over its assigned virtual paths in the same manner as the other nodes on the ring. A failed, or otherwise inoperative, node is removed from the ring without disrupting the traffic on the ring by essentially reversing the node insertion operations. After learning of the node failure, the hub node instructs the operative nodes to tear down the virtual path and associated virtual circuits that originate from or end at the failed node. Thereafter, the hub node directs the nodes to update their ring topology information to remove the failed node. In the meantime, the nodes continue to maintain the other virtual path and virtual circuit connections over the ring without disruption. When the failed node is later re-booted, the system follows the steps discussed above to re-insert the node, without disrupting the traffic on the ring.

System for non-disruptive insertion and removal of nodes in an ATM sonet ring

A unidirectional ring operates in accordance with a call setup procedure in which a single node on the ring, referred to herein as the &ldquo;hub node,&rdquo; simultaneously (i) processes call setup information from and provides call setup information to an ATM network, and (ii) controls the call setup operations of the nodes on the ring. When a source node that is part of the ring initiates a call, the node directs a call setup message to the hub node. The hub node determines if there is sufficient bandwidth available on the ring for the call. If so, the hub node passes the call setup message to the ATM network, and simultaneously multicasts call setup information to the nodes on the ring. While the call setup message is progressing over the ATM network in a conventional manner, the nodes on the ring simultaneously set up the associated virtual path and/or virtual circuit and send back connection information to the hub node. The hub node monitors the connection information to ensure that the connection over the ring is established. If the call is ultimately rejected by the destination node, the hub node multicasts to the nodes on the ring an instruction directing them to tear down the associated virtual circuit and/or virtual path. When the hub node receives a call setup message which is directed to a destination node that is a ring node, the hub node determines if there is sufficient bandwidth available on the ring to handle the call. If so, the hub node sends an appropriate acknowledgment message back to the node that sent the message to the ring. The hub node also multicasts call setup information to the nodes on the ring, instructing them simultaneously to set up a virtual circuit and/or virtual path for the call. If the destination node rejects the call, the hub node sends an appropriate call rejection message over the network to the source node and instructs the ring nodes to tear down the associated virtual circuit and/or virtual path.

Rapid call establishment in ATM rings

A system with predefined primary and secondary routes sets up corresponding virtual circuits over both routes. A destination node maintains primary and secondary sets of routing tables that contain, respectively, routing information over each of the rings for every virtual circuit. A destination node disables the appropriate entries in the routing tables for the non-selected route, such that the node ultimately discards the traffic received over that route. To switch a virtual path, and thus all of the virtual circuits included therein, from the selected route to the nonselected route, the destination node disables and enables the appropriate entries in the primary and the secondary routing tables, to switch from using the set-up virtual circuits on the selected route to using the set-up virtual circuits on the non-selected route. To detect an event that triggers protection switching, such as path failure, the system uses &ldquo;Continuity OAM cells&rdquo; to provide path status information to the nodes. A source node periodically originates Continuity OAM cells over the primary and the secondary routes. A node that receives a Continuity OAM cell multicasts the cell to (a) the processor on the node and (b) without alteration to successive nodes on the same ring. If the node processor determines that the node has not received a predetermined number of Continuity OAM cells from the source node over the selected route within a selected number of time windows, the node determines that there is a path failure on the selected route. The affected destination node then triggers protection switching if, in the same windows, the node receives an appropriate number of Continuity OAM cells from the source node over the non-selected route. To facilitate detection of path degradation, the source nodes include error counts for the associated ring interface in the Continuity OAM cells. A destination node detects path degradation when the accumulated error rate over a selected route from a source node exceeds a predetermined maximum. The destination node then triggers protection switching if the accumulated error rate over the non-selected route is better than that over the selected route by a predetermined margin.

Rapid ring protection switching system

Growing community of inventors

Brighton, MA, United States of America

Bappa Sinha