Abstract: In one embodiment, a method includes identifying at a node, a destination for which a primary path is defined between the node and the destination, transmitting a request message to a neighbor node for routing information for a backup path between the node and the destination, receiving a reply message containing an address identifying the backup path, and building a backup tunnel to the address. The request message includes at least one node identified as a node to be avoided in the backup path and a request cost. An apparatus for building backup tunnels is also disclosed.

Abstract: In one embodiment, a method includes identifying at a node, a destination for which a primary path is defined between the node and the destination, transmitting a request message to a neighbor node for routing information for a backup path between the node and the destination, receiving a reply message containing an address identifying the backup path, and building a backup tunnel to the address. The request message includes at least one node identified as a node to be avoided in the backup path and a request cost. An apparatus for building backup tunnels is also disclosed.

Abstract: Techniques for recovering lost routes include receiving reported costs for transmitting data to a destination from neighboring nodes; and determining total costs as a sum of costs for transmitting data packets to the neighboring nodes and a corresponding reported cost. A selected neighboring node with a minimum total cost is determined as the next hop for the route to the destination. A feasible successor set of neighboring nodes which have reported costs less than the total cost of the selected neighboring node and excluding the selected neighboring node, and successor data about the feasible successor set, are determined. The successor data is sent to the neighboring nodes. A neighboring node that loses a route to the particular destination node is able to determine whether to query the sending node while recovering a lost route to the destination based on the successor data, thereby reducing network resource consumption.

Abstract: A system for optimizing the performance of a network. In an illustrative embodiment, the system includes a memory containing one or more data structures data structures. A first module determines transit-only information. A second module selectively omits the transit-only information from the one or more data structures. In a more specific embodiment, the one or more data structures include one or more tables. The transit-only information includes interface Internet Protocol (IP) addresses associated with network interfaces that are transit-only interfaces. The route information may include network topology information. The one or more tables may include a network topology table. Plural routers positioned in the network incorporate one or more instances of the system. The routers may employ one or more routing protocols, such as include Enhanced Interior Gateway Routing Protocol (EIGRP) or Routing Information Protocol (RIP).

Abstract: A static neighbor configured network device is configured with a static neighbor feature including a static neighbor list of network devices. The static neighbor configured network device is in a network segment that includes a dynamically configured network device. The static neighbor configured network device receives a multicast packet from the dynamically configured network device. It is determined if the dynamically configured network device is included in the static neighbor list of the static neighbor configured network device. The multicast packet is accepted if the dynamically configured network device is found in the neighbor list for the static neighbor configured network device. Also, adjacency for the static neighbor configured network device is not lost with any other currently adjacent network devices when the multicast packet is received.

Abstract: Techniques for sending data in a packet-switched communications network include determining multiple subsets of adjacent network nodes of the network. The adjacent network nodes communicate without intervening network nodes with a particular network node through an interface on the particular network node. Each subset includes multiple adjacent network nodes. Subset definition data is sent through the interface. The subset definition data indicates which adjacent network nodes belong to which subset. Data for fewer than all adjacent network nodes in all subsets are sent by including, in a multicast data packet sent over the interface with a multicast destination address, subset identifier data that indicates a particular subset. When such data is received by a node, it is discarded unless the subset identifier matches the receiving node's subset identifier. Among other effects, this allows routing messages to be more efficiently sent to better performing neighboring network nodes.

Abstract: Techniques for recovering lost routes include receiving reported costs for transmitting data to a destination from neighboring nodes; and determining total costs as a sum of costs for transmitting data packets to the neighboring nodes and a corresponding reported cost. A selected neighboring node with a minimum total cost is determined as the next hop for the route to the destination. A feasible successor set of neighboring nodes which have reported costs less than the total cost of the selected neighboring node and excluding the selected neighboring node, and successor data about the feasible successor set, are determined. The successor data is sent to the neighboring nodes. A neighboring node that loses a route to the particular destination node is able to determine whether to query the sending node while recovering a lost route to the destination based on the successor data, thereby reducing network resource consumption.

Abstract: A first router is configured for monitoring prescribed attributes of an active path connected to the first router, and supplying an update message to a second router, according to a prescribed routing protocol such as Enhanced Interior Gateway Routing Protocol (EIGRP), that specifies a detected change by the first router in at least one of the prescribed attributes of the connected active path. Hence, the second router, in response to receiving the update message, can update an internal topology table based on the detected change in the active path connected to the first router, and selectively adjust an internal routing table based on the detected change relative to queuing policies for prescribed data flows.