摘要:
In general, techniques are described for summarizing label mappings and thereby enabling longest-prefix match within Multi-Protocol Label Switching (MPLS) networks. More specifically, a first router included within a first area of a network comprises a control unit that maintains a label space defining labels available for mapping to a plurality of addresses assigned to network devices within the network. The control unit reserves a contiguous set of the labels of the label space and maps the contiguous set of labels to first area addresses. The first area addresses include those addresses of the plurality of addresses available for assignment to network devices within the first area. The first router also includes an interface card that transmits, to a second router of a second area of the network, an advertisement that advertises a summarized version of the mapping between the contiguous set of labels and the first area addresses.
摘要:
Principles of the invention are directed to techniques for allowing a router forwarding packets within a computer network to perform two or more forwarding tree decisions per memory access. The router may implement forwarding information in the form of a radix tree having a number of nodes, and received packets may contain keys identifying a packet destination. The router may traverse the tree by testing two or more path control bits within the key per each of the traversed nodes. The values of the path control bits in the key determine the path traversed along the tree. The router also stores intermediate bit patterns at each node and tests intermediate bits in the key to determine whether a particular node is the best match to the routing prefix contained in the key, thereby eliminating a need to backtrack up the tree.
摘要:
A layer 2 transport network, and components thereof, supporting virtual network functionality among customer edge devices. Virtual private network configuration can be accomplished with merely local intervention by preprovisioning extra channel (or circuit) identifiers at each customer edge device and by advertising label base and range information corresponding to a list of channel (or circuit) identifiers.
摘要:
A layer 2 transport network, and components thereof, supporting virtual network functionality among customer edge devices. Virtual private network configuration can be accomplished with merely local intervention by preprovisioning extra channel (or circuit) identifiers at each customer edge device and by advertising label base and range information corresponding to a list of channel (or circuit) identifiers.
摘要:
Techniques are describe for establishing an overall label switched path (LSP) for dynamic load balancing of network traffic being sent across a network using the a resource reservation protocol such as Resource Reservation Protocol with Traffic Engineering (RSVP-TE). The tunnel may be a single RSVP-TE Label Switched Path (LSP) that is configured to automatically and dynamically load balance network traffic across different sub-paths of the RSVP-TE LSP over the network. The ingress device of the overall multi-path LSP can analyze traffic statistics to determine when a network traffic demand differs from a currently reserved bandwidth of the overall multi-path LSP by at least a threshold amount, and can automatically add or remove a sub-path from the overall multi-path LSP to adjust capacity of the overall multi-path LSP to correspond to the currently reserved bandwidth.
摘要:
Principles of the invention relate to techniques for transmission of Layer 2 (L2) traffic over a point to multi-point (P2MP) label switched path (LSP) within a multi-protocol Label Switching (MPLS) network. A source or ingress network device may implement the techniques to connect multiple L2 interfaces to a P2MP LSP within an MPLS network via a P2MP Pseudo-Wire (PW) mechanism that emulates delivery of L2 data units over a packet switched network, such as the MPLS network. The ingress network device first establishes the P2MP LSP and then connects two or more L2 interfaces to the P2MP LSP via the P2MP PW mechanism. Egress network devices may also implement the techniques to terminate the P2MP LSP and de-multiplex traffic arriving via multiple P2MP PWs carried over the same P2MP LSP.
摘要:
Techniques are described for verifying a status of a set of paths through a computer network for two or more connectivity protocols. For example, a node uses a first connectivity protocol to concurrently learn information that will cause packets conforming to the first connectivity protocol and packet conforming to a second connectivity protocol to traverse a set of paths through a computer network. After learning this information, the node may verify a status of each of the paths using the first connectivity protocol. In addition, the node may verify a status of each of the paths using the second connectivity protocol. By verifying the status of the paths using both the first and the second connectivity protocols, the node may be able to quickly and accurately determine whether a path has failed.
摘要:
When a node has to restart its control component, or a (e.g., label-switched path signaling) part of its control component, if that node can preserve its forwarding information across the restart, the effects of such restarts on label switched path(s) include the restarting node are minimized. A node's ability to preserve forwarding information across a control component (part) restart is advertised. In the event of a restart, stale forwarding information can be used for a limited time before. The restarting node can use its forwarding information, as well as received label-path advertisements, to determine which of its labels should be associated with the path, for advertisement to its peers.
摘要:
A network device includes a memory, a routing engine and a forwarding engine. The memory stores a forwarding table and the routing engine constructs a first composite next hop that includes multiple next hops, where each of the multiple next hops represents an action to be taken on a data unit as it transits the network device or represents another composite next hop, and where the first composite next hop specifies a function to be performed on the plurality of next hops. The routing engine further stores the composite next hop in an entry of the forwarding table. The forwarding engine retrieves the composite next hop from the forwarding table, and forwards a data unit towards one or more network destinations based on the composite next hop.
摘要:
A method and a network device for enabling communication between unnumbered interfaces are provided. A device level address may be assigned to a network device. The network device may announce the assigned device level address to a neighboring network device over a link. A corresponding device level address associated with the neighboring network device may be received over the link. A route may be stored including the received device level address associated with the neighboring network device and the link. In some implementations, the announcement of the assigned device level address is performed during protocol configuration.