Various embodiments provide a system for modifying a channel binding in order to relay packets between a relay client and a peer in a peer-to-peer (P2P) communication event across a network. A relay server receives a request to bind a channel in order to relay the packets for the communication event. The relay server creates requirements for a communication path. The relay server sends the requirements to a Software Defined Networking (SDN) controller. The SDN controller in turn creates and installs flows and flow tables in SDN switches to relay the packets across the network for the communication event.

The present disclosure is directed towards systems and methods for dynamic routing on an IP address shared by a cluster of nodes. In an implementation, a first node of a cluster of nodes can receive a unicast routing protocol packet from a peer router. The unicast routing protocol packet can be addressed to a shared IP address established across the cluster of nodes. The cluster of nodes can be intermediary to a plurality of clients and a plurality of servers. The first node can identify a second node identified as a routing leader. The first node can steer the packet to the second node in response to determining that the routing protocol packet is a unicast routing protocol packet. The second node can be configured to advertise virtual IP address routes to the network over the routing adjacency and maintain routing adjacencies.

A packet transmission method includes: receiving, by a first network device, a first packet sent by a previous-hop device of the first network device, where the first packet includes an SR header; generating, by the first network device, cache index information of the SR header, and storing the cache index information and the SR header; generating, by the first network device, a second packet based on the first packet, where the second packet includes the cache index information but does not include the SR header; and sending, by the first network device, the second packet to a second network device.

Techniques are disclosed for session-based routing within Open Systems Interconnection (OSI) Model Layer-2 (L2) networks extended over Layer-3 (L3) networks. In one example, L2 networks connect a first client device to a first router and a second client device to a second router. An L3 network connects the first and second routers. The first router receives, from the first client device, an L2 frame destined for the second client device. The first router generates an L3 packet comprising an L3 header specifying L3 addresses of the first and second routers, a first portion of metadata comprising L2 addresses for the first and second client devices, and a second portion of metadata comprising L3 addresses for the first and second client devices, and forwards the L3 packet to the second router. The second router recovers the L2 frame from the metadata and forwards the L2 frame to the second client device.

Methods and system for directing traffic flows to a fast data path or a slow data path are disclosed. Parsers can produce packet header vectors (PHVs) for use in match-action units. The PHVs are also used to generate feature vectors for the traffic flows. A flow training engine produces a classification model. Feature vectors input to the classification model result in output predictions predicting if a traffic flow will be long lived or short lived. The classification models are used by network appliances to install traffic flows into fast data paths or the slow data paths based on the predictions.

Methods and system for directing traffic flows to a fast data path or a slow data path are disclosed. Parsers can produce packet header vectors (PHVs) for use in match-action units. The PHVs are also used to generate feature vectors for the traffic flows. A flow training engine produces a classification model. Feature vectors input to the classification model result in output predictions predicting if a traffic flow will be long lived or short lived. The classification models are used by network appliances to install traffic flows into fast data paths or the slow data paths based on the predictions.

The disclosure provides for a system that includes a network controller. The network controller is configured to receive information from nodes of a network, where nodes include one node that is in motion relative to another node. The network controller is also configured to generate a table representing nodes, available storage at each node, and possible links in the network over a period of time based on the information, and determine a series of topologies of the network based on the table. Based on received client data including a data amount, the network controller is configured to determine flows for the topology. The network controller then is configured to generate a schedule of network configurations based on the flows, and send instructions to the nodes of the network for implementing the network configurations and transmitting client data.

A bulk material tender includes a mobile frame, a hopper, and a discharge system. The mobile frame has a left side and a right side. The hopper is disposed on the mobile frame. The discharge system is configured to discharge particulate matter from the hopper. The discharge system includes a discharge auger, a deploying actuator, and a positioning actuator. The discharge auger presents a proximal end and a distal end. The deploying actuator is configured to selectively emplace the discharge auger in a stowed orientation and a deployed orientation, wherein the distal end is adjacent to the hopper in the stowed orientation. The positioning actuator configured to selectively emplace the discharge auger along the left side and the right side of the mobile frame. Once emplaced, the discharge auger discharges particulate material from the hopper toward a target location.

Some embodiments provide a novel method for deploying different virtual networks over several public cloud datacenters for different entities. For each entity, the method (1) identifies a set of public cloud datacenters of one or more public cloud providers to connect a set of machines of the entity, (2) deploys managed forwarding nodes (MFNs) for the entity in the identified set of public cloud datacenters, and then (3) configures the MFNs to implement a virtual network that connects the entity's set of machines across its identified set of public cloud datacenters. In some embodiments, the method identifies the set of public cloud datacenters for an entity by receiving input from the entity's network administrator. In some embodiments, this input specifies the public cloud providers to use and/or the public cloud regions in which the virtual network should be defined. Conjunctively, or alternatively, this input in some embodiments specifies actual public cloud datacenters to use.

To determine whether to a delay a routing update, a router generates graphs of the network topology before and after the failure of a link or node and compares the graphs to identify changes in the network topology. Based on an analysis of the changes, the router determines whether to implement, continue and/or abort a preconfigured delay for making routing updates.