An example method for a device to implement one of the nodes in a wireless network for processing packets includes submitting a request to a network-management system in the network to become a node in the network, after having registered with the network-management system, determining neighboring nodes, flooding the wireless network with a link-state advertisement, the link-state advertisement providing neighboring relationships of the node, constructing switching rules for the node based on a tree switching network portion of the network, processing the packets received by the node with the switching rules, the switching rules defining at least one of (1) an ingress link to a parent node with a power capability greater than the node and (2) egress links to child nodes with a mobility greater than the node, and in response to having determined a failed link to a neighboring node, informing a node at the end of an ingress wireless link and the network-management system of the failed link.

In one embodiment, a first node in a network receives one or more bitmaps from one or more child nodes of the first node according to a directed acyclic graph (DAG). Each of the one or more child nodes is associated with a corresponding unique bit position in the one or more bitmaps. The first node stores, in a forwarding table, the one or more bitmaps received from the one or more child nodes of the first node. The first node receives a message that includes a destination bitmap that identifies one or more destinations of the message via one or more set bits at bit positions associated with the one or more child nodes. The first node forwards the message towards the identified one or more destinations based on the destination bitmap and the one or more bitmaps stored in the forwarding table of the first node.

An example system in accordance with an aspect of the present disclosure includes an inspection engine and a forwarding engine. The inspection engine is to identify whether a tunneled network packet is associated with the multicast group address and a VNI that is contained in the mapping table. The forwarding engine is to forward or discard the packet in response to whether the VNI is contained in the mapping table.

In some examples, method includes receiving, with a software-defined network (SDN) controller in an SDN containing a plurality of controlled network nodes, a dynamic network parameter for the SDN from a controlled network node in the SDN, selecting, with the SDN controller, an adjusted spanning tree protocol (STP) path cost value for a path cost along a datapath between a source network node and a destination network node in the SDN based on the received dynamic network parameter, and installing, with the SDN controller, the adjusted STP path cost value on controlled network nodes along the datapath.

Examples relate to fast failover recovery in software defined networks. In some examples, a failure in a first primary tree is detected during data transmission of a data packet, where the primary tree is associated with a first group entry that is configured to direct each of the data packets to one of a first set of destination devices. A notification of the failure is sent to a remote controller device, where the remote controller device identifies backup trees of the route trees that does not include the failure. After the remote controller device updates the first group entry to be associated with a first backup tree that minimizes congestion, each of the data packets are sent to one of a second set of destination devices that are associated with the first backup tree.

A controller assigns variable length addresses to addressable elements that are connected to a network. The variable length addresses are determined based on probabilities that packets are addressed to the corresponding addressable element. The controller transmits, to the addressable elements via the network, a routing table indicating the variable length addresses assigned to the addressable elements. Routers or addressable elements receive the routing table and route one or more packets over the network to an addressable element using variable length addresses included in a header of the one or more packets.

In one embodiment, a multiple spanning tree (MST) region is defined in a network, where the MST region includes a plurality of network nodes interconnected by links. A MST cluster is defined within the MST region, where the MST cluster includes a plurality of network nodes selected from the plurality of network nodes of the MST region. A network node of the MST cluster generates one or more MST bridge protocol data units (BPDUs) that present the MST cluster as a single logical entity to network nodes of the MST region that are not included in the MST cluster, yet enables per-multiple spanning tree instance (per-MSTI) load balancing of traffic across inter-cluster links that connect network nodes included in the MST cluster and network nodes of the MST region that are not included in the MST cluster.

A node receives status data associated with a current collector in the network, where the node is active on the current collector. The node also receives status data associated with a candidate collector in the network, where the node is not active on the candidate collector. An analysis of the status data of the collectors is generated, where the analysis includes at least comparing respective network loads reported in the received status data. An optimal collector is determined from among the current collector and the candidate collector. The determination of the optimal collector is based at least in part upon the analysis of the status data of the collectors. The node remains active on the current collector when the current collector is determined to be the optimal collector, and the node becomes active on the candidate collector when the candidate collector is determined to be the optimal collector.

In one embodiment, a learning data processor determines a plurality of machine learning features in a computer network to collect. Upon receiving data corresponding to the plurality of features, the learning data processor may aggregate the data, and pushes the aggregated data for select features to interested learning machines associated with the computer network.

A distributed system for collecting and processing packet routing information is provided. A service provider, such as a content delivery network service provider, can maintain multiple Points of Presence (“POPs”). Routing computing devices associated with each POP can forward information about the packet routing information to a packet routing management component. The packet routing component can process the information provided by the various POPs. The packet routing component can then update, or otherwise modify, packet routing information used by one or more of the POPs. Accordingly, the packet routing management component can then selectively distribute the updated or modified packet routing information, including the distribution to all POPs, the targeted distribution to specific POPs and the creation of centrally accessible routing information.