Systems and techniques, including special messages and state machines, are described that configures an intermediate site to dynamically trigger creation of and removal of a dynamic conduit between two sites based on usage that is tracked at the sites. The intermediate site providing WAN-to-WAN forwarding between the two sites, monitors throughput statistics on each local WAN link (LWL) associated with the two sites. If traffic between the two sites passes a configured first threshold or if LWL usage passes a configured second threshold, the intermediate site sends a message to the two sites to set up a dynamic conduit directly coupling the two sites. Busy lists are used to keep track of eligible site pairs. Once a dynamic conduit is set up between two sites, a grow technique tests the dynamic conduit increasing communication flows between the two sites each configured sampling period before putting the conduit in normal use.

Provided is a machine-implemented method for operating a device in a network, comprising entering the device into a non-router membership relation with a self-organizing subnet of the network; receiving at least one message making known parameters of the subnet, the at least one message comprising a centrally-generated and propagated sequence number; detecting, based on the received at least one message comprising a sequence number, a deterioration in propagation of said at least one message; and responsive to the detecting, entering the device into a router membership relation with the self-organizing subnet of the network.

Systems and methods provide congruent bidirectional Segment Routing (SR) tunnels, namely congruent and fate-shared traffic forwarding for bidirectional SR tunnels. A bidirectional SR tunnel, as described herein, includes two unidirectional SR tunnels where the forward and reverse traffic directions follow the same path through the network when forwarded based on prefix and adjacency Segment Identifiers (SIDs). The term “congruent” is used herein to refer to the fact that the two unidirectional SR tunnels, i.e., the forward and reverse traffic directions, follow the same path through the network but in opposite directions. The guarantee of congruency is based on modification of the Segment Identifier (SID) configuration at the source nodes of each tunnel. Accordingly, the present disclosure maintains compatibility with existing Segment Routing configurations with the modifications solely at the source nodes.

A method for connecting graph processing cut endpoints is disclosed. The method comprises determining a first capability matrix comprising capability values, applying weights to the capability values of the first capability matrix resulting in a first weighted capability matrix, and determining a second capability matrix comprising, for each pairing of another side of the cut endpoint, capability values. For both, the first capability matrix and second capability matrix, the capability values relate at least to a platform, a runtime framework, a connection technology, a protocol for a connection of a node of the stream processing system. The method comprises further determining a joint weighted capability matrix for two corresponding cut endpoints of a cut by multiplying corresponding capability matrix values of the first weighted capability matrix and the second capability matrix, and selecting a preferred connection for the cut endpoints.

An illustrative embodiment disclosed herein is an apparatus including a processor having programmed instructions to determine that a first packet, received from a first VM on a first host, has a destination address associated with a second VM on the first host, send the first packet to a service-focused bridge, generate a rule to redirect the first packet to the second VM based on the destination address, and generate execute the rule in response to receiving the first packet from the service-focused bridge.

Methods and apparatus for efficiently servicing isochronous streams (such as media data streams) associated with a network. In one embodiment, an Isochronous Cycle Manager (ICM), receives multiple independent streams of packets that include isochronous packets arriving according to different time bases (e.g., where each stream has a different time base). The packets are sorted by the ICM into a buffering mechanism according to their required presentation time. Additionally the ICM calculates a launch time for each packet. The NIC transmits the packets from the queue according to an access scheme, such as a time division multiplexed (TDM) scheme where each of a plurality of cycles is subdivided into time slots. During appropriate time slots, the NIC transmits the packets in chronological order, as read out of the buffering mechanism.

Techniques for identifying service paths for satisfying a service demand in a data communication network are disclosed. Aspects include identifying a plurality of vertices in a communications route cycle, the communications route cycle alternating through the plurality of vertices and a plurality of edges in a sequence; identifying a plurality of internal paths in the communications route cycle, each of the plurality of internal paths is disjoint to the plurality of edges; identifying a first internal path and a second internal path among the plurality of internal paths as a crossing pair of internal paths; detecting a failure of at least two edges among the plurality of edges; and identifying, based on the communications route cycle and the identified crossing pair of internal paths, a service path that satisfies the service demand in response to detecting the failure of the at least two edges.

A system and method for facilitating communication between one or more of a plurality of user virtual machines and external devices is disclosed. The system includes a plurality of uplink bridges configured to facilitate communication between the plurality of user virtual machines and the external devices, a plurality of local bridges, with each of the plurality of user virtual machines being connected to one or more of the plurality of local bridges, and a first multiplexing bridge connected to the plurality of local bridges. The first multiplexing bridge is configured to direct data between the plurality of user virtual machines and the plurality of uplink bridges. The system also includes a second multiplexing bridge connected to the first multiplexing bridge and the plurality of uplink bridges. The second multiplexing bridge is configured to direct the data between the first multiplexing bridge and the plurality of uplink bridges.

Systems and techniques are described for a path maximum transmission unit (MTU) discovery method that allows the sender of IP packets to discover the MTU of packets that it is sending over a conduit to a given destination. The MTU is the largest packet that can be sent through the network along a path without requiring fragmentation. The path MTU discovery method actively probes each sending path of each conduit with fragmentation enabled to determine a current MTU and accordingly increase or decrease the conduit MTU. The path MTU discovery process is resilient to errors and supports retransmission if packets are lost in the discovery process. The path MTU discovery process is dynamically adjusted at a periodic rate to adjust to varying network conditions.

Examples relate to hierarchical switching devices comprising a plurality of sub-switches forming a fully interconnected all-to-all network. The sub-switches comprise internal input ports and internal output ports to exchange packets with other sub-switches within the fully interconnected all-to-all network. The internal input ports of the sub-switches have exclusive access to a queue partition for each external output port of the respective sub-switch. A switch controller receives a packet at a first sub-switch of the plurality of sub-switches that is to be routed to a particular external output port of a second sub-switch of the plurality of sub-switches. The switch controller routes the packet directly from the first sub-switch to the second sub-switch using an internal output port of the first sub-switch and a queue partition of the second sub-switch that is for the particular external output port of the second sub-switch.