Abstract: A fair weighted-hashing technique may be used in load balancing among a group of modules. In one implementation, a device may maintain a table that relates how incoming client resource requests are to be distributed among the modules. The device may update the table, in response to an indication that an additional module, associated with a module identifier, is to be included in the group of modules. The updating may include determining a number of entries to add to the table for the additional module, calculating a first hash value for each of the number of entries, and modifying the table by writing the module identifier to one or more sequential entries of the table, beginning at an index into the table corresponding to the first hash value.

Abstract: Techniques are described for reducing unnecessary upstream traffic toward a rendezvous point (RP) of a network using Protocol Independent Multicast Bidirectional Mode. The techniques include determining whether the RP needs to receive multicast traffic for a multicast group and, when the RP does not need to receive the multicast traffic, sending RP-prune control messages for the multicast group to downstream routers. Upon receiving an RP-prune control message, a downstream router may prune an outgoing interface for the multicast group to prevent the downstream router from forwarding multicast traffic for the multicast group toward the RP. The downstream router may terminate the RP-prune control message or propagate the RP-prune control message to a further downstream router. Routers that send RP-prune control messages also create or update a RP-prune state for the multicast group that includes interfaces for the downstream routers to which the RP-prune control messages were sent.

Abstract: A router for switching data packets from a source to a destination in a network in which the router includes a distributed memory. The distributed memory includes two or more memory banks. Each memory bank is used for storing uniform portions of a data packet received from a source and linking information for each data packet to allow for the extraction of the uniform portions of a data packet from distributed locations in memory in proper order after a routing determination has been made by the router.

Abstract: Multicast traffic received by a subnet that uses IGMP/PIM snooping may be efficiently processed so that only required multicast router interfaces are used. A router may, for example, receive a source-specific PIM join/prune message indicating that a multicast receiver of the multicast traffic is to join/leave a multicast group to receive/stop traffic from a multicast source; determine whether the router is a first hop router relative to a subnet of the multicast source; and forward, when the router is a first hop router relative to the subnet of the multicast source and is a non-designated router, the source-specific PIM join/prune message towards the subnet.

Abstract: Managing TCP anycast requests at content delivery network nodes is disclosed. In some embodiments, serving a request includes receiving a request at a node of a plurality of nodes comprising a content delivery network, wherein each of the plurality of nodes share a same anycast IP address to which the request is directed and servicing the request at the node.

Abstract: A transport LAN segment service is provided over a transport network. Responsibilities for configuring, provisioning and forwarding over a transport LAN segment are divided between layer 2 and 3 service provider edge devices, where the layer 3 edge device handles discovery and tunneling responsibilities, the layer 2 edge device handles learning and flooding responsibilities, and information can be exchanged between the layer 2 and 3 edge devices. Configuration is simplified by advertising TLS-label information, layer 2 address learning, and flooding when the needed configuration information has not yet been learned or discovered.

Abstract: Techniques for facilitating the operation of one or more virtual networks are described. In some examples, a system may include a first controller node device configured to control operation of a first set of elements in the one or more virtual networks, wherein the first set of elements includes a first server device. The system may also include a second controller node device configured to control operation of a second set of elements in the one or more virtual networks, wherein the second set of elements includes the second server device. The first controller node device and the second controller node device are peers according to a peering protocol by which the first controller node device and the second controller node device exchange information relating to the operation of the first set of elements and the second set of elements.

Abstract: In general, techniques are described for providing thread-level tracing within an operating system kernel. A computing device comprising processes, an operating system and a control unit that executes the operating system may implement the techniques. Each of the processes comprises an instance of an application program. Each of the processes comprises one or more separate threads of instruction execution. The operating system concurrently executes the processes. The control unit traces execution of a system call issued to the operating system by one of the threads currently being executed by the control unit to generate a trace log that traces execution of the system call by the operating system without tracing execution of any other ones of the one or more threads corresponding to those of the plurality of processes that are also being executed by the operating system during execution of the system call.

Abstract: In one embodiment, an apparatus includes a switch core that has a multi-stage switch fabric physically distributed among a set of chassis. The multi-stage switch fabric has a set of input buffers and a set of output ports. The switch core can be configured to be coupled to a set of edge devices. The apparatus can also include a controller implemented in hardware without software during operation and with software during configuration and monitoring. The controller can be coupled to the set of input buffers and the set of output ports. The controller can be configured to send a flow control signal to an input buffer from the set of input buffers when congestion at an output port from the set of output ports is predicted and before congestion in the switch core occurs.

Abstract: A system selectively drops data from queues. The system includes a drop table that stores drop probabilities. The system selects one of the queues to examine and generates an index into the drop table to identify one of the drop probabilities for the examined queue. The system then determines whether to drop data from the examined queue based on the identified drop probability.

Abstract: A layer 2 transport network, and components thereof, supporting virtual network functionality among customer edge devices. Virtual private network configuration can be accomplished with merely local intervention by preprovisioning extra channel (or circuit) identifiers at each customer edge device and by advertising label base and range information corresponding to a list of channel (or circuit) identifiers.

Abstract: In general, techniques are described for determining a physical network path taken by packets of a network packet flow. The techniques may be applied to determine, or “trace,” a physical network path in the virtualized network domain. In some examples, a network device includes one or more processors and a switch executed by the processors to forward packets of a packet flow to a physical network path. The network device also includes a flow trace module to generate one or more flow trace packets having incrementally increasing respective time-to-live (TTL) values, wherein the switch module forwards the flow trace packets on an outbound interface of the network device for the physical network path, and wherein the flow trace module receives corresponding time exceeded messages for the flow trace packets, wherein each of the time exceeded message includes a source network address of a network element on the physical network path.

Abstract: In general, techniques are described for enhancing the Application-Layer Traffic Optimization (ALTO) service to supplement network topological grouping with location-based groupings to account for endpoint mobility. For example, as described herein, an ALTO server maintains physical location information for a network of one or more endpoints that provides a service. A PID generator of the ALTO server aggregates the endpoints into a set of one or more PIDs based at least on the physical location information for the endpoints, wherein each PID is associated with a subset of the endpoints. The ALTO server generates network and cost maps for the ALTO service that include PID entries to identify a respective subset of the endpoints associated with each of the set of PIDs and cost entries that incorporate cost that reflect physical distances among endpoints.

Abstract: An upgrade procedure may be executed at a network device to allow the network device to automatically upgrade an operating system to a new operating system. The upgrade procedure may include receiving a bootable image file of an upgrade operating system that is different from an operating system being run by the network device and which uses a file system format that is different from the file system format of the operating system being run by the network device. The bootable image file includes a binary file format that conforms to the file system format used by the upgrade operating system. The network device allocates contiguous space on a drive of the network device and copies the bootable image file to the contiguous space. The network device also modifies a master boot record of the drive such that the contiguous space on the drive is reflected in the master boot record as a bootable partition of the drive.

Abstract: A device receives distances between an access point, located on a floor of a building, and other access points located on the same floor, and determines, based on the distances, relative location information associated with the access point, where the relative location information provides a location of the access point relative to the other access points. The device also determines, using a triangulation method, an actual location of the access point based on the relative location information. The device further maps the actual location of the access point to a floor plan of the floor, and displays the floor plan with the mapped actual location of the access point.

Abstract: A method and apparatus stores media content in a variety of storage devices, with at least a portion of the storage devices having different performance characteristics. The system can deliver media to a large number of clients while maintaining a high level of viewing experience for each client by automatically adapting the bit rate of a media being delivered to a client using the client's last mile bit rate variation. The system provides clients with smooth viewing of video without buffering stops. The client does not need a custom video content player to communicate with the system.

Abstract: A system may include a power module that includes a group of power supplies, particular ones of the group of power supplies being operable at a group of voltages ranging from a first voltage to a second voltage. The system may further include a controller coupled to the particular ones of the group of power supplies, the controller being to ramp up an output voltage, associated with the group of power supplies, from the first voltage to the second voltage in a group of discrete steps; where ramping up the output voltage by a particular one of the group of discrete steps is performed while a load is receiving power from the group of power supplies; and where ramping up the output voltage by a particular one of the group of discrete steps prevents over-current protection on the group of power supplies from being activated.

Abstract: A device may store a first and second queue of packets, calculate an average queue size based on the number of packets in the first and second queues and discard a packet when the packet is a session creation packet and the calculated average queue size is greater than a threshold value.

Abstract: An endpoint integrity system controls access to resources of a protected network for endpoint devices attempting to access the protected network. The system may include a number of evaluation modules that communicate with an endpoint device. The evaluation modules generate policy results for the endpoint device, in which each of the policy results assume one of three or more states, called a multi-state policy result. The multi-state policy results are combined to produce a combined Boolean policy result.

Abstract: A rack system may include a first plurality of line cards, where a particular one of the first plurality of line cards receives or sends packets via ports; a plurality of fabric cards, where a particular one of the plurality of fabric cards includes a switching fabric; a second plurality of line cards, where a particular one of the second plurality of line cards receives or sends packets via ports; a first backplane that connects the first plurality of line cards to the plurality of fabric cards; and a second backplane that connects the second plurality of line cards to the plurality of fabric cards.