Abstract: In one embodiment, a packet switching device sends packets to be sent from a single link of a bundled link interface to multiple egress network processing units (on a same or different line cards). A single one of these multiple egress network processing units is configured to be in the active mode sending particular sets of packets. The other egress network processing units are configured for these particular sets of packets to be in the non-active mode, and hence, will correspondingly drop these particular sets of packets. In case of failure, an egress network processing unit can quickly (e.g., changing a flag) be changed to the active mode to quickly reduce or eliminate loss of packets.

Abstract: In one embodiment, backbone service instance identifiers (I-SIDs) of backbone frames are modified based on flow identification of frames encapsulated therein to induce entropy into the headers of the backbone frames. Backbone packet switching devices use the modified service instance identifier to load balance the corresponding frame through the backbone network. At an exit point of the backbone network, the original backbone service instance identifier (I-SID) associated with a frame encapsulated in a backbone frame is recovered from the modified service instance identifier, with this recovery typically including determining the flow identification of the frame encapsulated in the backbone frame.

Abstract: In one embodiment, a packet switching device creates multiple virtual packet switching devices within the same physical packet switching device using virtual machines and sharing particular physical resources of the packet switching device. One embodiment uses this functionality to change the operating version (e.g., upgrade or downgrade) of the packet switching device by originally operating according to a first operating version, operating according to both a first and second operating version, and then ceasing operating according to the first operating version. Using such a technique, a packet switching device can be upgraded or downgraded while fully operating (e.g., without having to reboot line cards and route processing engines).

Abstract: In one embodiment, a device receives a first packet stream and a second packet stream over different paths through a network, wherein each of said sent first and the second packet streams includes a same replicated stream of packets. The apparatus processes packets of the first packet stream when the first packet stream is in an active packet stream, and while buffering and subsequently dropping packets of the second packet stream when the second packet stream is in a non-active state. In response to identifying a difference in a number of packets in the same replicated stream of packets received in the second packet stream compared to in the first packet stream equaling or exceeding a predetermined threshold, the second packet stream becomes in the active state and missing packets are forwarded from the buffered second stream packets.

Abstract: In one embodiment, micro-loops are avoided in ring topologies of packet switching devices by changing the order of propagation of link state information concerning failed communications between a particular packet switching device and a neighbor packet switching device. In one embodiment, the particular packet switching device communicates link state information of a high cost of the particular communications (e.g., in the direction from particular to neighbor packet switching devices) such that this link state information will propagate towards the particular packet switching device from at least from the furthest packet switching device in the ring topology that is currently configured to forward packets having a destination address of the neighbor packet switching device through the particular packet switching device.

Abstract: In one embodiment, a packet switching device is configured to convert an Internet Protocol Version 6 (IPv6) destination address, of a received particular IPv6 packet, to a second, shorter destination address. This second destination address is then used to determine forwarding information for the received IPv6 packet, which is forwarded accordingly. In one embodiment, this second address is a 32-bit address, and in particular, an Internet Protocol Version 4 (IPv4) address. Thus, one embodiment can use the IPv4 forwarding infrastructure of a packet switching device for determining how to forward IPv6 packets. In a network according to one embodiment, packets are encapsulated in an IPv6 packet using an IPv6 destination address (that can be converted to an IPv4 address) of an egress edge packet switching device. Thus, core packet switching devices can forward IPv6 packets using IPv4 lookup operations.

Abstract: In one embodiment, a Bidirectional Forwarding Detection (BFD) asynchronous mode session is established between two packet switching devices interconnected by one or more physical links. Prior to L2 or L3 services being established, each of these packet switching devices does not know the Media Access Control (MAC) nor Internet Protocol (IP) addresses of each interface of the other packet switching device that is connected to one of these link(s). A request to establish a BFD session is sent from one packet switching device to the other, with a MAC frame including the request being addressed to a group, broadcast, or other address that the receiving packet switching device will recognize and thus process the received request. Based on information contained in this received MAC frame, the receiving packet switching device has the information it needs, and sends a BFD control frame to the other packet switching device.

Abstract: In one embodiment, an Internet Protocol (IP) routing information base of a packet switching device is filtered to produce a significantly smaller subset of IP routes that are installed in one or more forwarding information bases for forwarding of IP packets. In one embodiment, these smaller forwarding information bases are located in memory local to a network processor to more quickly perform lookup operations thereon. In one embodiment, one or more of these forwarding information bases is used only for exact matching of addresses (not longest prefix matching). In one embodiment, the IP routes in these smaller forwarding information bases substantially correspond to packet switching devices in a network (e.g., core and edge routers), such as in contrast to including all the IP routes for devices external to the network.

Abstract: In one embodiment, a logical packet switching device has its switching fabric extended between multiple physical devices, such as, but not limited to, over one or more networks (e.g., over tunnel(s), point-to-point link(s), and/or public and/or private L2 or L3 network(s)). In particular, one embodiment extends the switching fabric between multiple different physical devices by effectively merging, at least from the perspective of ingress and/or egress line cards, a switching fabric in each of these multiple different physical devices. In this regard, an ingress lookup operation in a first physical device of one embodiment produces information which is used by the switching fabric in a different physical device to forward a packet to the appropriate egress line card in the different physical device.

Abstract: In one embodiment, protocol violations of a particular protocol are induced at one or more predetermined intervals within a particular stream of information encoded according to the particular protocol in order to produce a marked particular stream of information for use in subsequent identification of the marked particular stream of information. The marked stream is multiplexed or otherwise communicated to a second device. The second device detects, and typically corrects, the induced protocol violations. And based on which stream of information included the induced protocol violations and the multiplexing/distribution pattern of the other streams of information, the second device can identify which stream is which and process or forward accordingly.

Abstract: In one embodiment, a programmable priority encoder is configured to receive inputs, including an ordered list of a plurality of input request values each representing either a request or a non-request, and a starting position within the ordered list of the plurality of input request values. The programmable priority encoder is configured to generate an identification of a result position of a first input indicating said request in order from a position identified from the starting position within the ordered list. In one embodiment, the programmable priority encoder includes a hierarchal structure of logic blocks including a plurality of columns of logic blocks; wherein a first-stage column of the plurality of columns of logic blocks is configured to operate on at most N input values; and wherein the ordered list of the plurality of input request values consists of N input request values.

Abstract: In one embodiment, a first device communicates with a second device, including sending and receiving one or more link bundle control packets and/or link control packets over a link bundle. The link bundle includes a plurality of links with each link being coupled between the first device and the second device for communicating information. These link bundle control packets are directed at the operation of the link bundle in its entirety; while the link control packets which are directed to the operation of a corresponding single particular link. The first device communicates with the second device, including operating the link bundle according to these one or more link bundle control packets and/or link control packets.

Abstract: In one embodiment, forwarding information bases (FIBs) are selectively populated in a packet switch. A packet switching device determines, based on one or more protocol signaling messages, a subset, which is less than all, on which FIBs a lookup operation may be performed for identifying forwarding information for a received particular packet. The packet switching device populates each of these FIBs, but not all of the FIBs of the packet switching device, with forwarding information corresponding to the particular forwarding value. Thus, FIB resources are consumed for only those FIBs which could actually be used, and not all of the FIBs, for forwarding packets in the data plane of the packet switching device, whether these packets are received on a primary or backup path.

Abstract: In one embodiment, line cards of packet switching or other network devices are configured for terminating pseudowires. Typically, this includes multiple line cards being configured for terminating a same pseudowire, which allows the corresponding pseudowire traffic to be received by any one of these multiple line cards. Each of these pseudowire-terminating line cards is typically configured to apply one or more features to a pseudowire packet. Examples of these features include, but are not limited to: Access Control List, Quality of Service, Netflow, and Lawful Intercept. For a received packet to be sent out one of these pseudowires, a two-stage lookup operation can be used to first identify the pseudowire over which to forward the packet; and a second lookup operation based on the pseudowire to identify forwarding information corresponding to a path through a network over which a corresponding pseudowire is configured.

Abstract: In one embodiment, single-homing and active-active multi-homing is provided in a Virtual Private LAN Service (VPLS). A customer edge node actively communicates frames of a same Virtual Private Network (VPN) instance with two or more VPLS nodes of a VPLS network. The VPLS nodes are configured to appropriately forward frames throughout the VPLS network: without looping of a frame sent by the same external node back to the same external node, without flooding multiple copies of a frame to the same external node, and while performing learning of addresses in forwarding tables of said VPLS nodes such that said forwarding tables of said VPLS nodes converge despite frames of the same LAN service being received by said at least two of said VPLS nodes from the same external node.

Abstract: One embodiment includes, inter alia, methods, apparatus, computer-storage media, mechanisms, and/or means associated with automated transitioning between different communication protocols in a network. In one embodiment, automatic transition routers are automatically discovered along with the knowledge of what non-native protocols need to be transported across a network. Communication pathways are automatically established as needed to transport these non-native protocols. One embodiment is particularly useful in transitioning a network from one protocol to another, such as from Internet Protocol version 4 to version 6.

Abstract: In one embodiment, a packet switching device assigns a same particular packet switching label to each particular route of a plurality of particular routes having the same one or more best paths, wherein the plurality of particular routes includes routes from at least two different forwarding groups. A forwarding group is defined as a specific route, one or more routes associated with a same customer edge router, or one or more routes associated with a single virtual routing and forwarding domain (VRF). The packet switching device advertises to other packet switching device(s) to add this same particular label to packets having one of the plurality of particular routes, which they do. The packet switching device then packet switches packets based on the particular label received in a label field in a header of these packets.

Abstract: In one embodiment, packet flows are distributed among groups, such as, but not limited to, queues or links. For example in the context of a bundled interface in which multiple links appear as a single logical interface on each of the sender and receiver, packet flows are distributed among these multiple links by the sender. When one or more links become unavailable, packet flows of the affected links are reassigned to other active links, while packet flows assigned to the unaffected links remain associated with the same link in contrast to prior systems which do not attempt to preserve prior associations between packet flows and links. By maintaining these associations, the receiver of the packets does not need to adjust to the different arrival links of packet flows.

Abstract: Data path processing information is included in the pseudowire layer of pseudowire packets in order to provide information for use in the data path processing of data (e.g., a packet), typically, but not always, included in the payload of the pseudowire packet itself. The pseudowire packet typically includes in corresponding fields: a pseudowire label for identifying a pseudowire type; a pseudowire control word; and payload data. The pseudowire type identifies the structure of the pseudowire control word field and the payload field, including the location of data path meta data, such as in the pseudowire control word field or payload field. This data path meta data identifies one or more attributes for use in processing the payload data.

Abstract: An application node advertises service(s), using a label distribution protocol, that it offers to other network nodes and a corresponding label to use to identify these services(s). For example, a Targeted Label Distribution Protocol (tLDP) session may be established between a packet switching device and the application node providing these services to communicate the advertisement. Packets are encapsulated and sent from a service node (e.g., packet switching device) with the corresponding label to have one or more advertised services applied to the packet by an application node (e.g., a packet switching device and/or computing platform such as a Cisco ASR 1000).