59 results about "Link state packet" patented technology

Methods and apparatus are provided for optimizing the reintroduction of a network node into a network. Information about neighboring nodes is stored in persistent memory. The network node can then be reinitialized and reintroduced into the network. Upon reintroduction, the network node can transmit heartbeat messages such as Hello messages to its neighboring nodes using information stored in persistent memory. A link state packet request message such as a Complete Sequence Numbers Packet referencing dummy link state information is transmitted to a neighboring node. A partial packet request message such as a Partial Sequence Numbers Packet referencing the dummy link state packet from the neighboring node can acknowledge that the Complete Sequence Numbers Packet has been received.

The present invention discloses an efficient architecture for routing in a very large autonomous system where many of the layer 3 routers are attached to a common connection-oriented layer 2 subnetwork, such as an ATM network. In a preferred embodiment of the invention, a permanent topology of routers coupled to the subnetwork is connected by permanent virtual circuits. The routers can further take advantage of both intra-area and inter-area shortcuts through the layer 2 network to improve network performance. The routers pre-calculate shortcuts using information from link state packets broadcast by other routers and store the shortcuts to a given destination in a forwarding table, along with corresponding entries for a next hop along the permanent topology. The present invention allows the network to continue to operate correctly if layer 2 resource limitations preclude the setup of additional shortcuts.

A technique enables an intermediate network node to efficiently process link-state packets using a single running context (i.e., process or thread). The intermediate network node floods received link-state packets (LSP) before performing shortest path first (SPF) calculations and routing information base (RIB) updates. In addition, the node limits the number of LSPs that are permitted to be flooded before the node performs its SPF calculations. More specifically, if the number of link-state packets that are flooded during a flooding cycle exceeds a first predetermined threshold value, the node performs the SPF calculations before additional packets may be flooded. The intermediate network node also limits how long its RIB update may be delayed in favor of flooding operations. When the number of LSPs flooded after the SPF calculations exceeds a second predetermined threshold value or there are no more packets to be flooded, the node updates the contents of its RIB based on the SPF calculations.

In a link state protocol such as an interior gateway protocol (IGP), link state advertisements or link state packets (LSA/LSPs) are sent only to network nodes that have expressed interest in them, rather than always flooding them.

In one embodiment, a method for setting up a flow-through mesh group (FTMG) for transmitting link-state packets (LSPs) in a network having a plurality of nodes interconnected by links. The FTMG is a combination of multiple spanning trees for the network through which LSPs are forwarded. FTMG set-up messages are received at ports of each node of the network from peer ports of linked nodes. FTMG set-up messages identify root nodes of the multiple spanning trees and the transmission modes of the peer ports. The FTMG set-up messages are used to determine (1) a root node for each spanning tree, (2) a root port on each node for each spanning tree, and (3) directionality of ports of the nodes. FTMG set-up messages are then used to determine the transmission mode of ports of the nodes and, subsequently, to update the spanning trees and transmission modes, as needed.

According to a specific embodiment of the present invention, methods and apparatus are disclosed for generating network topology information. A first link state packet associated with a first node is identified. A second link state packet associated with a second node is identified. Network topology information associated with the first node is generated using information from the first link state packet. Verification of two-way connectivity between the first node and the second node is deferred until analysis of the second link state packet is initiated for generating network topology information associated with the second node.

A method and apparatus are disclosed for performing a shortest path first network routing path determination in a data communications network based in part on information about links that are associated as shared risk link groups. Micro-loops are avoided in computing shortest path first trees by considering whether links are within shared risk link groups. In a first approach, for each link state packet in a link state database, listed adjacencies are removed if the link between the node originating the LSP and the reported adjacency belongs to a shared risk link group for which one component (local link) is known as down, and a shortest path first computation is then performed. In a second approach, during the SPT computation and after having added a first node to a path, each neighboring node is added to a tentative tree if and only if, a link between the first node and the neighboring node does not belong to a shared risk link group for which one component (local link) is known as down.

We formulate the network-wide traffic measurement/analysis problem as a series of set-cardinality-determination (SCD) problems. By leveraging recent advances in probabilistic distinct sample counting techniques, the set-cardinalities, and thus, the network-wide traffic measurements of interest can be computed in a distributed manner via the exchange of extremely light-weight traffic digests (TD's) amongst the network nodes, i.e. the routers. A TD for N packets only requires O(loglog N) bits of memory storage. The computation of such O(loglog N)-sized TD is also amenable for efficient hardware implementation at wire-speed of 10 Gbps and beyond. Given the small size of the TD's, it is possible to distribute nodal TD's to all routers within a domain by piggybacking them as opaque data objects inside existing control messages, such as OSPF link-state packets (LSPs) or I-BGP control messages. Once the required TD's are received, a router can estimate the traffic measurements of interest for each of its local link by solving a series of set-cardinality-determination problems. The traffic measurements of interest are typically in form of per-link, per-traffic-aggregate packet counts (or flow counts) where an aggregate is defined by the group of packets sharing the same originating and/or destination nodes (or links) and/or some intermediate nodes (or links). The local measurement results are then distributed within the domain so that each router can construct a network-wide view of routes/flow patterns of different traffic commodities where a commodity is defined as a group of packets sharing the same origination and/or termination nodes or links. After the initial network-wide traffic measurements are received, each router can further reduce the associated measurement/estimation errors by locally conducting a minimum square error (MSE) optimization based on network-wide commodity-flow conservation constraints.

The invention discloses a method for avoiding neighborhood flap in an intermediate system (IS). When updating the state information of N links in the link state data base (LSDB) of an IS, the IS does not send a neighboring IS a partial sequence number packet (PSNP) carrying the summary information of the N link state packets (LSPs) one-to-one corresponding to the state information of the N links instead of the N LSPs, so that the neighboring IS can update the LSDB thereof according to the summary information of the N LSPs; wherein N is natural numbers. The invention also discloses an IS. The technical proposal provided in the invention remarkably reduces ISIS packet when updating the LSPs, so as to avoid ISIS neighborhood flap due to fullness of ISIS packet sending and receiving queues.

Disclosed are methods and apparatus for flooding link state packets or packet data units (LSP's) from a first node to one other LSP receiving node, where the LSP's are flooded in an order that is related to relative costs of various paths emanating from the first node. That is, the first node sends its own LSP first since the first node has no cost relative to itself. The first node then sends other LSP's that are associated with a plurality of other LSP sending nodes that have previously flooded their LSP's to the first node, and these other LSP's are sent by the first node to the other LSP receiving node in an order related to the relative costs associated with the paths from the first node to each of the other LSP sending nodes which had previously sent the other LSP's to the first node.

A technique efficiently avoids transient routing disturbances in link state routing protocols with fragmented link state packets (LSPs) in a computer network. According to the novel technique, a link state router (LSR) specifies which of two or more links are to be advertised in each of two or more corresponding LSP fragments. The LSR advertises the states of the specified links in the corresponding LSP fragments to one or more other LSRs. In other words, each link of the LSR is assigned to a particular LSP fragment, and the state of the link is always to be advertised in that particular LSP fragment (i.e., no fragment wrapping). Upon receiving the LSP fragments, the other LSRs may update the correct link states based on the individual LSP fragments, i.e., without transient routing disturbances caused by fragment wrapping.

The invention discloses a method for generating and distributing link state protocol data unit fragment messages, which comprises the following steps of: taking a standby main control board as a current active main control board in the restart process of an active main control board of equipment, generating the LSP (Link State Packet) fragment messages according to recalculated route information as well as a relation between TLV ID (Threshold Limit Value Identification) and LSP fragment numbers before the restart of the existing active main control board and only sending to neighbor equipment if compared with LSP fragment messages before the restart of the existing active main control board, the generated LSP fragment messages are changed. The invention also provides the equipment; and after the restart of the existing active main control board of the equipment is finished, the sending of the unnecessary LSP fragment messages is omitted.

The embodiments of the invention provide a link state synchronization method and a device which are applied to intermediate systems. The method comprises: sending solidified link state packets (LSP) to intermediate systems in neighboring relationship with itself wherein the solidified link state packets are link state packets (LSP) that are not subjected to aging; determining a targeted intermediate system wherein the targeted intermediate system is one that does not send partial sequence numbers packets (PSNP) to itself in a preset period; and sending the solidified link state packets (LSP) to the targeted intermediate system. The embodiments can avoid the periodical flooding LSP to all intermediate systems to maintain LSDB synchronization, therefore, saving network band width.

The invention discloses a method for protecting and recovering a cross-domain end-to-end label switched path and relates to the field of an automatic switched optical network. The method comprises the following steps of: S1, building a work path and N protection paths, wherein N is greater than or equal to 1; S2, judging the path in which a fault occurs, if the path is the work path, carrying out protection switching on business, and judging the position at which the fault is at the same time; if the position is in a source domain, a host domain or an intermediate domain, triggering one protection path bypassing a fault point in the domain in which the fault occurs; splicing the protection path with the protection paths without the fault in the other domains to form a new protection path; if the position is between the domains, performing no triggering on recovery of the work path with the fault; and if the fault is on the protection path, needing no protection switching on the business, triggering one new protection path in the domain, and splicing the protection path with the protection path without the fault in other domains to form another new protection path. According to the invention, by protecting, restoring and splicing an LSP (link state packet) of each segment, multipoint faults in the domains or between the domains are effectively resisted, and high reliability requirements of cross-domain business are met.

The invention provides a method and a device for preventing traffic from interrupting. According to the method and the device, a timer T1 and a function of keeping neighbor connection continuous are configured on routing equipment, when the configuration of an intermediate system to intermediate system routing protocol (IS-IS) of the routing equipment is changed, the connection between the routing equipment and current neighbors can be continuously maintained in a T1 time, and the update of link-state packets (LSPs) of the routing equipment and the current neighbors can also be ensured in the T1 time, and the traffic is continually to be forwarded for preventing the traffic from interrupting by utilizing the routing equipment and routers of the current neighbors.

The invention provides a route selection method and route selection device in a content center network. The route selection method comprises: a link state packet sent by each of a plurality of router nodes is received, wherein the link state packet sent by each router node at least includes identifier information of each router node, a link state between each router node and a neighbor node, and a cache loss probability of a cache packet in each neighbor node; a network topology structure of a content center network is constructed according to the link state information packets sent by the plurality of router nodes; and a shortest route for an expected cache hit path of a target cache data packet is calculated based on the network topology structure and the cache loss probability of the cache data packet of each neighbor node. Therefore, the user can obtain a request data packet rapidly.

A system comprising a plurality of nodes forming a network and a link-state packet capable of being transmitted by a source node in the network to a destination node in the network over a communication link; wherein the source node subtracts a transmission cost associated with the communication link from a time-to-live value associated with the link-state packet prior to transmitting the link-state packet to the destination node. Each node in the network may be configured to use a routing protocol, such as a no-sight routing protocol, to assign the transmission cost to the communication link.