A method of communicating between nodes in a network where a node receives a sequence of symbols that will form a packet on a first communications channel and has a planned packet that it would send on a second communications channel. A destination is encoded into an arbitration portion of a header sequence of the packet, the header sequence comprising a sequence of symbols. The transmission on the second communications channel is as per the planned packet, for as long as the symbols of the planned packet match the symbols being received on the first channel. An arbitration decision is made when the symbols do not match, with the node either continuing to send the rest of the planned packet, or the rest of the packet being received on the first communications channel.

A switch may operate in a cut-through mode and a store-and-forward mode. While in a default cut-through mode, the switch continuously monitors ports for certain health metrics. If those health metrics fall below a threshold, the switch changes to operate in a store-and-forward mode, either for a predetermined period of time or until the health metrics rise above a threshold, at which point the switch can resume cut-through mode operations. If health metrics fall below an even lower threshold, or remain below threshold for a predefined period of time, the switch can automatically alert a remote system or software process.

Methods and apparatus for controlling monitoring operations performed by various devices, e.g., access points, in a communications network and for using information obtained by the devices which perform the monitoring are described. The methods are well suited for use in a system with a variety of access points, e.g., wireless and/or wired access points, which can be used to obtain access to the Internet or another network. An access point, which has been configured to monitor in accordance with received monitoring configuration information, e.g. on a per access point interface basis, captures packets, stores captured packets, and monitors to detect communications failures corresponding to communications devices using said access point. In response to detecting a communications failure, the access point generates, an event failure notification indicating the type of detected failure and sends the event failure notification to the network monitoring node along with corresponding captured packets.

Systems and computer program products for performing retransmission of data packets over a network. A node receives a data packet with a source and a destination address. The data packet is sent along a network path to the destination address, and information associated with the data packet is sent to a controller node that is independent of the network path. A controller receives information associated with a data packet from any forwarder node within a plurality of forwarder nodes each monitoring communications along separate communications paths. An indication of a receipt acknowledgement for the data packet is received from a second forwarder node that is separate from the first forwarder node and the controller node. The receipt acknowledgement is correlated with the data packet and based on the correlating, data associated with retransmission processing of the data packet is deleted.

A receiving core reads a packet from an ingress interface, wherein the ingress interface is an interface corresponding to a forwarding group to which the receiving core belongs; the receiving core sends the read packet to a forwarding core in the forwarding group; and the forwarding core sends the packet to a corresponding egress interface.

A message transmission method, an access node, an access controller, and an access system, where in the method, a message is processed by an access controller instead of an access node such that the access node does not need to support a relay function, reducing operation and maintenance difficulties and costs. In the method, the access controller receives a first message from the access node, and the first message includes a first identifier. The access controller obtains a line identifier according to the first identifier. The access controller obtains a second message according to the line identifier, and the second message includes the line identifier. The access controller sends the second message to a relay server.

A transfer device includes an output-port decision unit to decide, on the basis of storage information stored in a frame input, an output port from which the frame is output from among a plurality of ports, an allocation unit to associate an input port to which the frame is input with an output port from which a frame which is transferred by a cut-through method is output on a one-to-one basis, and allocate a first frame to a first pathway transferring by the cut-through method and allocate a second frame to a second pathway transferring by a store-and-forward method on the basis of type information of an input port to which a frame has been input, class information of the frame, and the output port decided by the output-port decision unit, and an IET-output control unit to output the first frame from the output port, decide whether to divide the second frame on the basis of the class information of the second frame, and output the second frame from the output port on the basis of decision. Therefore, the transfer device can realize an IET low-latency transfer function by control simpler than that in the conventional transfer devices.

Disclosed are methods lossless traffic forwarding using distributed delay offset matching. The lossless traffic forwarding method includes calculating a delay offset between a first forwarding path between a transmitting node and a receiving node and a second forwarding path between the transmitting node and the receiving node, and controlling a buffer resource by an extent of the delay offset to delay packets to be forwarded on the first forwarding path.

A broker system dynamically adjusts data processing techniques used by a network gateway for forwarding data over a network from a client system to an exchange system based on a switch mode parameter. The broker system dynamically adjusts the network gateway from using an error-propagating data processing technique (e.g. cut-through switching) to a non-error-propagating data processing technique (e.g. store and forward switching) based on the switch mode parameter. In particular, the broker system may adjust the network gateway when a count of unsuccessfully validated protocol data units meets a threshold specified by the switch mode parameter. While using the error-propagating technique, the broker system may modify a portion of the data protocol unit corresponding to a transport layer of a network protocol used by the network when a protocol data unit is unsuccessfully validated.

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.