Abstract: In general, techniques described are for providing graceful restart procedures for network devices of label switched paths (LSPs) implemented with label stacks. For example, a restarting network device may include a processor coupled to a memory that executes software configured to: receive a path signaling message including a recovery object that defines a reverse path of the LSP from an egress network device of the LSP to the restarting network device, including at least an upstream label and a downstream label associated with the restarting network device; determine, based on the recovery object, the upstream label and the downstream label associated with the restarting network device; and instantiate a control plane state of the restarting network device based on the recovery object.

Abstract: In one example, a management component executes on a network device, a first routing component executing on a first server device physically separate from the network device to provide control plane routing functionality for a virtual node on the network device, and a second routing component executing on a second server device physically separate from the network device to provide control plane routing functionality for the virtual node, wherein the first routing component and the second routing component operate as a redundant master-backup pair for the virtual node. The management component is configured to detect first reachability of the network device with a first routing component and second reachability of the network device with a second routing component, and send a control message to each of the first routing component and the second routing component, the control message specifying reachability information indicating the first reachability and the second reachability.

Abstract: In general, the disclosure describes examples where a single software-defined network (SDN) controller establishes tunnels and controls communication on these tunnels between a plurality of virtual computing environments (VCEs). The SDN controller establishes the logical tunnel mesh to interconnect the plurality of VCEs in the multi-cloud network via respective connect gateway routers. To establish the logical tunnel mesh, the SDN controller is configured to determine one or more logical tunnels from the logical tunnel mesh to establish one or more communication links between a first VCE and a second VCE of the plurality of VCEs in the multi-cloud network. The SDN controller is configured to advertise the one or more logical tunnels to the first VCE and the second VCE.

Abstract: One or more devices are configured to receive information regarding network devices associated with a physical network. The one or more devices are configured further to generate configuration data based on the information regarding the network devices. The one or more devices are configured further to generate a virtual network based on the configuration data. The one or more devices are configured to send information regarding the virtual network to a client device. The one or more devices are configured to receive a change to the virtual network from the client device; and cause a change, corresponding to the change in the virtual network, to occur in the physical network.

Abstract: The techniques describe packet reordering for packets flowing on a new path in response to a change in internal forwarding paths in a network device. For example, a network device may dynamically change the selection of an internal forwarding path to achieve fabric path optimization (“OFP”) or to ensure optimized load balancing. Packets forwarded on the new path are buffered such that the transmission of packets forwarded on the new path are delayed for a buffering time period of at least the time in which a packet is being sent from the source packet processor to the initial destination packet processor.

Abstract: In one example, a network device comprising a first chassis of a multi-chassis link aggregation group (MC-LAG) having three or more chassis, comprises one or more network interfaces configured to receive a packet to be forwarded using the MC-LAG, and a control unit configured to determine whether the packet was received from a device outside of the MC-LAG, when the packet was received from the device outside of the MC-LAG, add data to the packet that identifies the first chassis as a source of the packet for the MC-LAG, and forward the packet via at least one of the network interfaces. In this manner, chassis of the MC-LAG can prevent forwarding of the packet to the source of the packet for the MC-LAG, based on the data that identifies a source of the packet for the MC-LAG.

Abstract: This disclosure describes techniques for presenting information about a network, virtualization infrastructure, cluster, or other computing environment, and may involve presentation of user interfaces that may enable nuanced, unique, and/or comprehensive insights into how infrastructure elements and computing resources are being used and information about patterns of usage and/or utilization. This disclosure also describes techniques for communicating, within a computing system, information used to create, update, and/or modify the user interfaces that present information about a network, virtualization infrastructure, cluster, or other computing environment.

Abstract: Techniques are described for avoiding traffic black-holing in a multi-homed Ethernet virtual private networks (EVPNs) in which a customer device (CE) is multi-homed to a plurality of multi-homing provider edge devices (PEs) via respective links of an Ethernet segment. An overlay network is created over the Ethernet segment, and the multi-homing PEs of the EVPN are configured with a common anycast IP address for respective virtual network interfaces. Upon election as active designated forwarder (DF) for the EVPN, the DF PE of the multi-homing PEs advertises toward the customer network an IGP metric for the anycast IP address that is lower than the IGP metric(s) advertised by any of the non-DF standby PE routers segment to direct the CE to forward network packets from the customer network to the DF PE over the respective link of the Ethernet segment.

Abstract: Photonically integrated normal incidence photodetectors (NIPDs) and associated in-plane waveguide structures optically coupled to the NIPDs can be configured to allow for both in-plane and normal-incidence detection. In photonic circuits with light-generation capabilities, such as integrated optical transceivers, the ability of the NIPDs to detect in-plane light is used, in accordance with some embodiments, to provide self-test functionality.

Abstract: A device may receive, from a first device, a port control protocol (PCP) request that includes a customer side translator (CLAT) prefix and one or more private internet protocol version X (IPvX) addresses. The PCP request may be received via an internet protocol version Y (IPvY) network. The device may store the CLAT prefix and the one or more private IPvX addresses using a data structure. The device may receive a packet that includes a private IPvX of the one or more private IPvX addresses and a private IPvY address that includes the CLAT prefix and a second instance of the private IPvX address. The device may use an application layer gateway (ALG). The device may translate the private IPvX address to a public IPvX address using the CLAT prefix. The device may provide the packet that includes the public IPvX address to a second device that supports IPvX.

Abstract: In one example, a merge point network device (MP) receives a plurality of resource reservation request messages for establishing a plurality of label switched paths (LSPs), wherein each of the plurality of LSPs has a common point of local repair network device (PLR) and has the MP as a common MP, wherein each of the resource reservation request messages identifies a common bypass tunnel that extends between the PLR and the MP and avoids a protected resource. The MP stores an association between the bypass tunnel and each of the plurality of LSPs. The MP receives a single message to trigger creation at the merge point network device of backup LSP state information for all of the plurality of LSPs. In response to receiving the single message, the MP installs state information for all of the LSPs that correspond to the bypass tunnel according to the stored association.

Abstract: A disclosed method may include (1) executing a virtual router that services traffic within a network in connection with a specific network consumer and (2) dynamically scaling memory of the virtual router to accommodate a networking need of the specific network consumer by (A) installing, in at least one component of a physical network device that hosts the virtual router, a set of networking objects that facilitate servicing the traffic in connection with the specific network consumer, (B) determining an amount of memory that is consumed by the set of networking objects at the component of a physical network device, and (C) modifying a configuration file of the virtual router such that the memory of the virtual router is scaled to store the set of networking objects via the component. Various other systems and methods are also disclosed.

Abstract: Techniques are described for detecting and correcting mis-programming of label information in a router of a label switched path (LSP) without initially triggering a tear-down of the LSP. For example, techniques described in this disclosure enable an ingress router to determine whether label information is correctly programmed between a routing engine (e.g., control plane) and a forwarding engine (e.g., forwarding plane) of a router in the LSP, and to correct any mis-programming of label information by informing the router to reprogram the forwarding engine with original forwarding label information associated with the LSP.

Abstract: In some examples, a method includes receiving, by a first network device, a private label route message from a second network device, the private label route message specifying a private label as a destination, a route distinguisher of an egress network device for the private label, a context protocol next hop address that identifies a private Multiprotocol Label Switching (MPLS) forwarding layer, and a next hop for the private label, determining, by the first network device and based on the private label route message, a label stack having a plurality of labels to use for forwarding traffic to the next hop for the private label, and storing, in a context forwarding table associated with the private MPLS forwarding layer, a private label destination with the label stack as a next hop for reaching the private label.

Abstract: The thermal impedance of p-i-n diodes integrated on semiconductor-on-insulator substrates can be reduced with thermally conducting vias that shunt heat across thermal barriers such as, e.g., the thick top oxide cladding often encapsulating the p-i-n diode. In various embodiments, one or more thermally conducting vias extend from a top surface of the intrinsic diode layer to a metal structure connected to the doped top layer of the diode, and/or from that metal structure down to at least the semiconductor device layer of the substrate.

Abstract: In some embodiments, an apparatus includes a memory and a processor operatively coupled to the memory. The processor is configured to send a stimulus signal at a frequency that corresponds to a first frequency value to a tributary channel of a coherent optical transponder. The processor is configured to adjust an amplitude of the stimulus signal and receive a first plurality of output optical power values. The processor is configured to adjust the frequency of the stimulus signal and receive a second plurality of output optical power values. The processor is configured to determine a bandwidth limitation and a modulation nonlinearity, and then send a first signal to a first filter to reduce the bandwidth limitation and a second signal to a second filter to reduce the modulation nonlinearity.

Abstract: In some examples, a method includes receiving, by a first ingress network device for a network, a source tree join route message from an egress network device for the network, specifying a multicast source and a multicast group, and in response to receiving the source tree join route message, determining, by the ingress network device, whether the multicast source is multi-homed to the network via the first ingress network device and a second ingress network device for the network. The method includes, in response to determining that the multicast source is not multi-homed, forwarding traffic for the multicast source on an inclusive provider tunnel without initiating setup of a selective provider tunnel to the egress network device, and in response to determining that the multicast source is multi-homed, initiating setup of a selective provider tunnel to the egress network device and terminating forwarding multicast traffic on the inclusive provider tunnel.

Abstract: In integrated optical structures (e.g., silicon-to-silicon-nitride mode converters) implemented in semiconductor-on-insulator substrates, wire waveguides whose sidewalls substantially consist of portions coinciding with crystallographic planes and do not extend laterally beyond the top surface of the wire waveguide may provide benefits in performance and/or manufacturing needs. Such wire waveguides may be manufactured, e.g., using a dry-etch of the semiconductor device layer down to the insulator layer to form a wire waveguide with exposed sidewalls, followed by a smoothing crystallographic wet etch.

Abstract: A device may receive information associated with a set of types of virtual network interface cards (vNICs). A hypervisor, of the device, being capable of connecting a vNIC, to a virtual bus, to connect a virtual machine, of the device, to a network. The device may connect the vNIC, to the virtual bus, based on the information associated with the set of types of vNICs. The vNIC being associated with a type of the set of types. The device may determine whether the virtual machine is compatible with the vNIC based on connecting the vNIC to the virtual bus. The device may selectively connect another vNIC, to the virtual bus, based on determining whether the virtual machine is compatible with the vNIC.

Abstract: In some embodiments, a network management module is operatively coupled to a set of edge devices that are coupled to a set of peripheral processing devices. The network management module can receive a signal associated with a broadcast protocol from an edge device from the set of edge devices in response to that edge device being operatively coupled to a switch fabric. The network management module can provision that edge device in response to receiving the signal. The network management module can define multiple network control entities at the set of edge devices such that each network control entity from the multiple network control entities can provide forwarding-state information associated with at least one peripheral processing device from the set of peripheral processing devices to at least one remaining network control entity from the multiple network control entities using a selective protocol.