Abstract: Disclosed are systems, apparatuses, methods, and computer-readable media for configuring network groups without software-based processing and management. A method includes: transmitting a topology request from a first leaf node to a first spine node; in response to receiving a network topology reply from a second leaf node, generating a network topology identifying a group of leaf nodes, the network topology reply including network information associated with the second leaf node, and in response to a request to send a message from a first network node connected to the first leaf node to a second network node connected to the second leaf node, transmitting the message to the second leaf node based on the network information from the network topology reply. In some aspects, the first spine node or a second spine node are configured to receive the message and transmit the message to the second leaf node.

Abstract: In-band network telemetry for multicast is provided by injecting a synthetic telemetry packet into the multicast. Each first node of each branch replicates the synthetic packet and passes on copy to the next branch and sends one copy down the nodes in the branch sequentially. The synthetic telemetry packet can include flags with have values, including, for example, whether each node should hold or send data collected by the node according to the synthetic data packet, how long a node is to hold collected telemetry data, and what data to collect. The node sends a postcard to the originator with the telemetry data according to the instructions in the synthetic telemetry packet.

Abstract: Presented herein are techniques to reduce the number of redirected subscriber packet flows while performing sticky hierarchical load balancing. An Nth head end network element may be activated such that a plurality of N head end network elements are active and capable of receiving and processing one or more packet flows. A primary load balancer may then be directed to overwrite a portion of pointers of a hash table in an evenly distributed manner with pointers to the Nth head end network element such that packet flows are forwarded to the Nth head end network element, wherein the hash table retains a static number of entries as the number of head end network elements is modified.

Abstract: In one embodiment, a packet switching device includes one or more host devices and a cascade of aggregation nodes. The aggregation nodes aggregate customer traffic and communicate it with the host device. Typically the aggregation nodes are remotely located from the host device. The host device may be connected to one or both ends of the cascaded topology of aggregation nodes. In one embodiment, the cascaded topology of aggregation nodes automatically configures itself using initiation packets. In one embodiment, the cascaded topology of aggregation nodes reacts to detected faults, such as by changing direction packet traffic is sent through the cascaded topology. By cascading aggregation nodes, in contrast to having each aggregation node connected to the host device via one or more point-to-point links, communications costs are decreased in one embodiment.

Abstract: Presented herein are techniques to reduce the number of redirected subscriber packet flows while performing sticky hierarchical load balancing. An Nth head end network element may be activated such that a plurality of N head end network elements are active and capable of receiving and processing one or more packet flows. A primary load balancer may then be directed to overwrite a portion of pointers of a hash table in an evenly distributed manner with pointers to the Nth head end network element such that packet flows are forwarded to the Nth head end network element, wherein the hash table retains a static number of entries as the number of head end network elements is modified.

Abstract: In one embodiment, a packet switching device includes one or more host devices and a cascade of aggregation nodes. The aggregation nodes aggregate customer traffic and communicate it with the host device. Typically the aggregation nodes are remotely located from the host device. The host device may be connected to one or both ends of the cascaded topology of aggregation nodes. In one embodiment, the cascaded topology of aggregation nodes automatically configures itself using initiation packets. In one embodiment, the cascaded topology of aggregation nodes reacts to detected faults, such as by changing direction packet traffic is sent through the cascaded topology. By cascading aggregation nodes, in contrast to having each aggregation node connected to the host device via one or more point-to-point links, communications costs are decreased in one embodiment.

Abstract: In one example embodiment, a method may include receiving a service advertisement message from an advertising node via a ring-based network, determining a service map based on the received service advertisement message, and transmitting data to the advertising node. The service advertisement message may include an address of the advertising node and a service ID for one or more services. The service map may indicate an association between the address of the advertising node and the one or more service IDs. The transmitting data to the advertising node may include transmitting data for at least one of the services to the advertising node based on the service map via the ring-based network.

Abstract: A method for providing inter-ring protection in shared packet rings includes identifying an active node which is connected to a ring interconnect node on the same ring as the active node and connected to a peer node on a different ring with a ring interconnecting link. When the active node is in active mode and receives notification of a failure of the ring interconnecting link or peer node, the active node sends a message to the ring interconnect node so that the ring interconnect node switches from standby mode to active mode. The active node is then changed to standby mode.

Abstract: Techniques for load balancing network traffic over virtual point-to-point data links at a data-link layer include determining a particular group of virtual point-to-point data links that reach a particular local area network of one or more network nodes. A particular data-link address for a particular node of the particular local area network is associated with each virtual point-to-point data link of the particular group. When a data packet is received that includes data that indicates the particular data-link address as a destination; one virtual point-to-point data link of the group is determined to balance traffic among the group. The data packet is sent only over the one virtual point-to-point data link. These techniques enable virtual point-to-point data links through two or more intermediate network nodes, such as a pair of provider edge nodes, to be bundled for load balancing purposes.

Abstract: A method for preventing loops in a ring topology of a computer network is disclosed. The method includes receiving a packet at a node on a shared packet ring and inserting a header containing an ID identifying the node at which the packet was received on the ring as an originating node. The packet is forwarded onto the ring and the ID is checked when received on a node. The packet is dropped if the ID identifies the node as the originating node.

Abstract: Techniques for load balancing network traffic over virtual point-to-point data links at a data-link layer include determining a particular group of virtual point-to-point data links that reach a particular local area network of one or more network nodes. A particular data-link address for a particular node of the particular local area network is associated with each virtual point-to-point data link of the particular group. When a data packet is received that includes data that indicates the particular data-link address as a destination; one virtual point-to-point data link of the group is determined to balance traffic among the group. The data packet is sent only over the one virtual point-to-point data link. These techniques enable virtual point-to-point data links through two or more intermediate network nodes, such as a pair of provider edge nodes, to be bundled for load balancing purposes.