Systems, methods, and computer-readable media relate to providing a network management service. A system is configured to request first network information from a first component of a network using a public IP address for the first component, wherein the first network information includes private IP addresses for a second component in the network and translate, based on a mapping information for a private IP address space to a public IP address space, the private IP address for a second component to a public IP address for the second component. The system is further configured to request second network information from the second component using the public IP address and provide a network management service for the network based on the second network information.

Systems and methods for supporting a single logical IP subnet across multiple independent layer 2 subnets in a high performance computing environment. A method can provide, at a computer including one or more microprocessors, a logical device, the logical device being addressed by a layer 3 address, wherein the logical device comprises a plurality of network adapters, each of the network adapters comprising a physical port, and a plurality of switches. The method can arrange the plurality of switches into a plurality of discrete layer 2 subnets. The method can provide a mapping table at the logical device.

A system and method of transferring cells through a switch fabric having a shared memory crossbar switch, a plurality of cell receive blocks and a plurality of cell transmit blocks. The system determines, based on a number of cells queued up in respective output buffers in the cell transmit blocks, output buffers in the cell transmit blocks that can receive cells on a low latency path. The cells transferred include first cells that can be transferred on the low latency path and second cells that cannot be transferred via the low latency path. The first cells are transferred via a bypass mechanism in shared memory to the output buffers. The second cells are transferred by writing the second cells to shared memory, reading the second cells from shared memory and transferring the second cells read from shared memory to the output buffers in the cell transmit blocks.

Presented herein are techniques that enable Ethernet Virtual Private Networks (EVPNs) to support use cases where either the Layer 2 or Layer 3 route associated with a virtual endpoint is different after the virtual endpoint migrates/moves to a different location. In particular, a networking device running an overlay network detects that a virtual endpoint has migrated on the overlay network from a first computing device to a second computing device. The networking device determines a modified Layer 2/Layer 3 route for the virtual endpoint at the second computing device, as well as a sequence number for association with the modified Layer 2/Layer 3 route. The sequence number is determined based on a sequence number associated with a parent Layer 2 route for the modified Layer 2/Layer 3 route.

A network box accepts packets from a network in ingress ports of a first tier of leaf switches, adaptively routes the packets from the leaf switches to a second tier of spine switches within the network box, statically routes the packets from the spine switches to designated egress ports of the leaf switches, and transmits the packets from the designated egress ports into the network.

Methods for assisting data forwarding during convergence in a multi-homed network are disclosed. In one aspect, a first leaf node is configured to detect when a second leaf node advertises a set of Ethernet segments which are local to the first leaf and advertise reachability information for the second leaf, indicating itself as a backup for the second leaf during convergence. A spine node that receives advertisement messages from such first and second leaf nodes programs its routing table to indicate the direct route to the second leaf as the primary path and the route to the second leaf via the first leaf as a backup path to forward encapsulated packets destined to the second leaf. Upon failure of the second leaf, when the spine node receives data packets destined to the second leaf, the spine node sends the packets to the first leaf instead of the second leaf.

An example system for satellite payload communications includes a digital channelizer and a regenerative communications subsystem (RCS). The digital channelizer includes a plurality of inputs for receiving a plurality of signals from a plurality of uplink beams and a plurality of outputs for outputting the plurality of signals. The RCS includes a plurality of inputs selectably coupled to the digital channelizer outputs to receive signals from selected ones of the digital channelizer outputs and a plurality of outputs selectably coupled to the digital channelizer inputs to transmit the processed signals to selected ones of the digital channelizer inputs. The RCS is configured to process selected ones of the plurality of signals to produce processed signals.

The expansion of a network by converting the network from a 2-stage folded Clos network to a 3-stage folded Clos network can be cost prohibitive. The system and methods described herein relate to a hybrid network topology. More particularly, the disclosure describes a hybrid topology having internal switches configured in a multidimensional topology configuration. Each of the internal switches are connected to the network hosts with a folded Clos topology.

In one example, a network device is configured as a leaf node of an interconnected topology that defines a plurality of network paths from the network device to each of a plurality of other leaf nodes of the interconnected topology. The network device includes one or more processing units configured to forward a first packet of a packet flow along a first network path of the plurality of network paths, after receiving a second packet of the packet flow, determine an inactivity interval for the packet flow that represents an amount of time between receipt of the first packet and receipt of the second packet, compare the inactivity interval to a threshold, and when the inactivity interval is greater than the threshold, forward the second packet along a second network path of the plurality of network paths, wherein the second network path is different than the first network path.

Many network protocols, including certain Ethernet protocols, include specifications for multiplexing using of virtual lanes. Due to skews and/or other uncertainties associated with the process, packets from virtual lanes may arrive at the receiver out of order. The present disclosure discusses implementations of receivers that may use multiplexer based crossbars, such as Clos networks, to reorder the lanes. State-based controllers for the Clos networks and state-based methods to assign routes in are also discussed.