A multilayer architecture for supporting variable networks is discussed. The architecture includes an application layer in communication with a simulation and network management layer and a coupling layer. The coupling layer interfaces with physical connections within a network device and the remaining layers; and wherein the simulation and network management layer sets one or more network parameters within the application layer.

In one embodiment, a plurality of PODs is formed in a software defined networking (SDN) fabric, each POD comprising a plurality of leaf nodes and connected to a plurality of spine nodes in a spine layer of the SDN fabric. One of the plurality of PODs is designated as a super POD and link state information is provided for the entire fabric to the super POD by sending northbound advertisements in the fabric to the super POD. A disconnection is identified between a leaf node in the SDN fabric and a particular one of the spine nodes in the spine layer, based on the link state information provided to the super POD. The disconnection is repaired between the leaf node and the particular spine node in the spine layer.

Techniques for providing a backup network path using a standby wide area network (WAN) link with reducing monitoring. Packet loss and latency metrics are monitored for network paths in an adaptive private network (APN) connecting a first user and a second user according to control traffic operating at a first control bandwidth for each network path. A determination is made that a first network path uses a standby WAN link, has packet loss and latency metrics indicative of a good quality state, and has at least one characteristic that identifies the first network path as a backup network path. The control traffic is then reduced for the backup network path to a second control bandwidth substantially less than the first control bandwidth. The backup network path is made active when the number of active network paths is less than or equal to a minimum number.

In one embodiment, a method is performed. A fat tree route miner (FT-RM) entity may be used to establish a control plane session with a first spine node in communication with a network. The FT-RM entity may identify a prefix that is unreachable by the first spine node. The FT-RM entity may instruct a spine node to disaggregate the prefix.

Provided is a flow control device for suppressing a consumption of a network bandwidth in multipoint-to-multipoint communication. The flow control device is a device for controlling a plurality of flow-processing devices that process a packet for predetermined communication. The flow control device include first means for calculating, for each of predetermined flow-processing devices located at an edge to an external network, a shortest path to another of the predetermined flow-processing devices, second means for determining, based on a number of times a link between flow-processing devices is included in the shortest path, a candidate link used in a communication path for the predetermined communication; and third means for calculating a candidate communication path for the predetermined communication, based on the determined candidate link.

The present disclosure provides a proactive method of prefix disaggregation in a network fabric when one or more communication failures are detected. In one aspect, a method includes determining, by a first node of a network fabric, a corresponding prefix disaggregation policy for at least one second node of the network fabric, the corresponding prefix disaggregation policy identifying one or more network prefixes that are inaccessible via the first node when at least one communication failure is detected in association with the first node; sending the corresponding prefix disaggregation policy to the second node; and causing the second node to implement the prefix disaggregation policy upon detecting the at least one communication failure.

Systems and methods for supporting inter subnet partitions in a high performance computing environment. In accordance with an embodiment, a fabric manager can define a range of P_Key values, among a plurality of P_Key values, as a inter subnet partition (ISP) P_Key range. The fabric manager can communicate this defined range of P_Key values to a number of subnets, via their subnet managers. The subnet managers in each subnet retain management over their subnets. As there is no central management that configures each side of inter subnet communication, subnet managers on within participating subnets can set up ISP membership, and then exchange information with the other subnet.

Systems and methods for supporting efficient virtualization in a lossless interconnection network. An exemplary method can provide, one or more switches, including at least a leaf switch, a plurality of host channel adapters, wherein each of the host channel adapters comprise at least one virtual function, at least one virtual switch, and at least one physical function, a plurality of hypervisors, and a plurality of virtual machines, wherein each of the plurality of virtual machines are associated with at least one virtual function. The method can arrange the plurality of host channel adapters with one or more of a virtual switch with prepopulated local identifiers (LIDs) architecture or a virtual switch with dynamic LID assignment architecture. The method can assign each virtual switch with a LID. The method can calculate one or more linear forwarding tables based at least upon the LIDs assigned to each of the virtual switches.

One embodiment of the present invention sets forth a technique for compressing a forwarding table. The technique includes selecting, from a listing of network prefixes, a plurality of network prefixes that are within a range of a subnet. The technique further includes sorting the plurality of network prefixes to generate one or more subgroups of network prefixes and selecting a first subgroup of network prefixes included in the one or more subgroups of network prefixes. The technique further includes generating a synthetic supernet based on the first subgroup of network prefixes.

The disclosure is related to a multi-gateway communication method. The method is applied to a wireless gateway system that includes multiple nodes including a number of gateways. A root gateway is provided among the gateways. When a gateway receives an uplink data from a node, the uplink data is transmitted to a backend server if the gateway is the root gateway. The uplink data is transmitted to a parent gateway if the gateway is not the root gateway. On the other hand, when the gateway receives downlink data from a node, it is determined if a destination node is recorded in a downlink routing table of a descendant gateway if gateway is not connected with the destination node but with the descendant gateway. If the destination node is recorded in the downlink routing table, the data is transmitted to the descendant gateway; otherwise, the downlink data is discarded.