Abstract: A system and method for routing network packets. A switch fabric connects a plurality of forwarding units, including an egress forwarding unit and two or more ingress forwarding units, each ingress forwarding unit forwarding network packets to the egress forwarding unit via the switch fabric. The egress forwarding unit includes a scheduler and an output queue. Each ingress forwarding unit includes a Virtual Output Queue (VOQ) connected to the output queue and a VOQ manager. The scheduler receives time of arrival information for network packets stored in the VOQs, determines, based on the time of arrival information for each network packet, a device resident time for the network packets stored in the VOQs, and requests, from one of the VOQs and based on the device resident times, the network packet with the longest device resident time.

Abstract: A device receives network data associated with a network that includes a peer network with network devices interconnected by links, and receives constraints associated with determining traffic assignments for the network. The device determines traffic and costs associated with the network based on the network data, and determines traffic assignments based on the traffic and the costs associated with the network, the constraints, and a model for the constraints. The device determines tunnel use based on the traffic assignments, and determines peer link use based on the tunnel use. The device determines costs associated with the traffic assignments, the tunnel use, and the peer link use for the network, and generates traffic plans based on the traffic assignments, the tunnel use, the peer link use, and the costs. The device causes one of the traffic plans to be implemented in the network by the network devices and the links.

Abstract: In some embodiments, an apparatus comprises a memory and a processor operatively coupled to the memory. The processor is configured to receive, from a forward error correction (FEC) decoder of an optical transponder, a first plurality of pre-FEC bit error rate (BER) values at a plurality of times to identify a degradation over a first transmission path. The processor is configured to determine, based on the first plurality of pre-FEC BER values, a signal pattern. The processor is configured to adjust, based on the signal pattern, a set of parameters including a first threshold and a second threshold. The processor is configured to send, in response to a second pre-FEC BER value exceeding the second threshold and being below the first threshold, a signal to trigger traffic rerouting to a second transmission path to reduce traffic loss due to the degradation over the first transmission path.

Abstract: Techniques are disclosed for providing an inter-autonomous system (inter-AS) service between virtualized entities of one autonomous system with external entities of a different autonomous system. For example, a controller (e.g., software defined networking (SDN) controller) may provide multi-hop exterior Border Gateway Protocol (eBGP) redistribution of virtual private networking (VPN) labels between endpoints of different autonomous systems, otherwise referred to as “inter-AS option C.” As described in this disclosure, the SDN controller may facilitate the exchange of appropriate routing labels between endpoints of different autonomous systems to enable forwarding of traffic between the different autonomous systems.

Abstract: Techniques are described for signaling a resource reservation point-to-multipoint (P2MP) label switched path (LSP) in a ring network. For example, the techniques include sending a single resource reservation request message without the ingress needing to send multiple resource reservation request messages, one for each leaf network device of the P2MP ring LSP. The resource reservation request message includes a leaf identification object from which network devices of a P2MP ring LSP may explicitly or implicitly identify themselves as prospective leaf network devices of the P2MP ring LSP. A network device determined to be a leaf network device may send a resource reservation response message back to the ingress network device of the P2MP ring LSP and send a modified resource reservation request message (if explicit) and/or the same resource reservation request message (if implicit) to a next hop network device along the P2MP ring LSP.

Abstract: A system for resolving next hops in a distributed manner includes a plurality of next hop resolution subsystems, each of the plurality of next hop resolution subsystems including (1) a next hop registry, (2) a partial-view tree storing a local instance of next hop resolution information, and (3) a full-view tree storing a local instance of next hop resolution information. The system (1) receives a request to resolve a next hop; (2) obtains a plurality of partial results using the partial-view trees of the next hop resolution subsystems; (3) select a best one of the plurality of partial results; and 4) replies to the request using the selected best one of the plurality of partial results. The full-view trees are updated such that their contents tend to converge.

Abstract: In general, the disclosure describes techniques for evaluating application quality of experience metrics over a software-defined wide area network. For instance, a network device may receive an application data packet of a data flow. In response to receiving the application data packet, the network device determines whether a packet size of the application data packet is represented in a reference data store. In response to determining that the packet size is not represented in the reference data store, the network device predicts, based on the reference data store, flow metrics for the packet size for each of a plurality of Wide Area Network (WAN) links. The network device selects a WAN link on which to send the application data packet based on the predicted flow metrics.

Abstract: Techniques are disclosed for a unified control plane in a nested cluster environment. In one example, an underlay orchestrator for a virtualized computing infrastructure is configured to provision, in an underlay cluster of one or more servers, an overlay cluster of one or more overlay nodes. Each of the overlay nodes is a workload of one of the servers and has a virtual network interface. Further, each server executes a virtual router and a virtual router agent for the virtual router for routing packets among virtual network interfaces of the overlay nodes executed by the server. A network controller is configured to configure virtual network sub-interfaces for workloads of the overlay nodes to enable communications among workloads executing in different overlay nodes. Each of the sub-interfaces is a virtual network sub-interface of one of the virtual network interfaces.

Abstract: A device may comprise security processing units (SPUs) including a SPU to receive a session request. The SPU may identify global counter information and update counter information. The global counter information may include a global counter identifier and a global counter value. The update counter information may include an update counter identifier and an update counter value. The SPU may identify a global limit associated with the global counter, may determine that the global limit has not been met, and may cause the session to be created. The SPU may create a modified global counter value. The SPU may create a modified update counter value. The SPU may determine that a local update message is required based on the modified update counter value, and may provide the local update message to another SPU. The local update message may include the global counter identifier and the modified global counter value.

Abstract: A first device may provide, periodically throughout a test session and to neighboring devices that are in a network with the first device, a message request for measuring network performance. The neighboring devices, upon receiving the request message, are to use a relay mechanism to determine network performance indicator (NPI) values. The first device may receive, from the neighboring devices and periodically throughout the test session, a response message that includes the NPI values. The first device may determine additional NPI values that measure the network performance between the first device and the neighboring devices. The first device may determine overall NPI values based on the NPI values and the additional NPI values. The first device may identify a preferred next-hop to one of the neighboring devices based on the overall NPI values, where the preferred next-hop is part of a preferred path through the network.

Abstract: A network device receives first objects defining first network data associated with an asynchronous distributed network and an asynchronous distributed network operating system and generates a natural name graph based on the first objects. The network device provides information describing the natural name graph to other network devices in the asynchronous distributed network and receives a second object defining second network data. The network device determines whether one of the first objects in the natural name graph is a duplicate of the second object and removes the one of the first objects from the natural name graph, to generate an updated natural name graph, when the one of the first objects in the natural name graph is a duplicate of the second object. The network device provides the updated natural name graph to the other network devices to prevent the other network devices from generating erroneous results.

Abstract: A system may receive network device configuration information associated with a network service. The system may determine multiple settings associated with the network service based on receiving the network device configuration information. The multiple settings may include a common setting and a device-specific setting. The system may generate a first network service model of the network service based on the multiple settings. The first network service model may include multiple nodes corresponding to the multiple settings. The system may re-configure one or more nodes associated with the first network service model, and may generate a second network service model of the network service. The system may generate a user interface template based on the second network service model and may provide the user interface template to a client device for display. The client device may allow a configuration of the multiple settings based on the user interface template.

Abstract: An article may include an optical transceiver package, which may include a photonics component mounted in the optical transceiver package. The photonics component may generate heat in an operational state. The optical transceiver package may include a sealed thermal chamber that maintains the photonics component between a lower predetermined working temperature and a higher predetermined working temperature. The sealed thermal chamber may include a material that exhibits a first thermal conductivity below a lower predetermined threshold temperature and a second thermal conductivity higher than the first thermal conductivity above an upper predetermined threshold temperature. A method may include retaining the generated heat to raise the photonics component above a lower predetermined working temperature, and conducting the generated heat away from the optical transceiver package to lower the photonics component below an upper predetermined working temperature.

Abstract: In general, techniques are disclosed to facilitate communicating within computer networks. For example, a layer three (L3) router including a service card and an interface card may be configured to perform the techniques. The interface card receives a query from a network that sources communications in accordance with a plurality of models. The query may specify a customer device and one of the sourced communications, and request that the service card select one of the models for the specified sourced communication and the specified customer device. The service card further stores data defining a profile for the one of the customer devices. The service card may also, in response to the query, analyze the profile data for the specified customer device to determine the selected one of the models for the specified sourced communication with respect to the specified customer device.

Abstract: Techniques are described for providing session-aware, stateful network services to subscriber packet flows. Devices within a service provider network direct subscriber packets along service chains. Each tunnel is established to direct traffic according a particular ordered set of network services for the corresponding service chain. An ingress device for the tunnels encapsulate the subscriber packets and embed opaque session cookies that each uniquely identifies a collection of packet flows of a subscriber session amongst other packet flows transported by a given service tunnel. Each service node need only identify the tunnel on which a tunnel packet was received and the session cookie embedded within the tunnel packet to uniquely associate the encapsulated subscriber packet with a subscriber session, without needing to further inspect the encapsulated subscriber packet, and to index or otherwise retrieve state and statistics required to enforce the network service the service nod is programmed to deliver.

Abstract: Techniques are described for supporting Fibre Channel over Ethernet (FCoE) link aggregation groups (LAGs) between a server and a data center switch in a data center. The techniques enable an access switch in the data center switch to assign class identifiers to each member link in an FCoE LAG between an FCoE node (Enode) of the server and the access switch. In this way, the access switch is able to redirect FCoE traffic from a Fibre Channel forwarder (FCF) of a storage area network (SAN) toward the Enode on the correct member link of the FCoE LAG. The techniques also enable scaling of FCoE initialization protocol (FIP) and FCoE sessions by installing FIP snooping filters that use on a source media access control (MAC) address hit determination in ingress filter processors (IFPs) of the access switch to avoid session limitations of virtual local area network (VLAN) filter processors (VFPs).

Abstract: Techniques are disclosed for an Ethernet Virtual Private Network (EVPN) Virtual Private Wire Service (VPWS) network with service interface-aware forwarding. In one example, a first network device signals to a second network device, using EVPN route advertisements, a multi-service service tunnel to transport network packets for a plurality of services. The services are identifiable by virtual local area network (VLAN) identifiers in the packets. The first network device is configured with a single transport interface for the service tunnel and the single transport interface is configured with respective service interfaces for the services. The first network device detects failure of a failed service interface of the service interfaces and outputs, in response to the failure, an EVPN route withdrawal message for the service tunnel that identifies the service corresponding to the failed service interface.

Abstract: In some embodiments, an apparatus includes a flow control module configured to receive a first data packet from an output queue of a stage of a multi-stage switch at a first rate when an available capacity of the output queue crosses a first threshold. The flow control module is configured to receive a second data packet from the output queue of the stage of the multi-stage switch at a second rate when the available capacity of the output queue crosses a second threshold. The flow control module configured to send a flow control signal to an edge device of the multi-stage switch from which the first data packet or the second data packet entered the multi-stage switch.

Abstract: Apparatus and methods described herein relate to an apparatus including a memory and a processor operatively coupled to the memory. The processor can receive a set of network management device packages associated with a set of network management devices. The processor can merge a set of management device schema commands from the set of network management device packages with a unified schema to produce unified schema information. The processor can receive a unified schema command signal based on the unified schema information, and can convert the unified schema command signal into a set of management device schema command signals using the unified schema. The processor can also send each management device schema command signal from the set of management device schema commands, to network management devices.