Abstract: A network device for processing data packets includes input logic, a register, route lookup logic and output processing logic. The input logic receives a data packet having at least one multi-protocol label switching (MPLS) tag, stores the packet and forwards a portion of the packet including the MPLS tag. The route lookup logic receives the portion of the packet, generates next hop information based on the MPLS tag and determines whether at least one MPLS tag is to be removed from the packet before the packet is to be output by the network device. The route lookup logic also modifies a start offset value in the register for each MPLS tag that is to be removed and forwards the start offset. The output processing logic receives the start offset and begins processing the data packet for output at a location identified by the start offset.

Abstract: A system processes packet data received in a number of incoming streams of variable speeds. The system includes an input interface, input logic, and one or more packet processors. The input interface receives the packet data and outputs the data using a first arbitration element. The input logic includes flow control logic, a memory, and a dispatch unit. The flow control logic initiates flow control on the data output by the input interface. The memory stores the data from the input interface. The dispatch unit reads the data from the memory using a second arbitration element. The packet processor(s) process the data from the dispatch unit.

Abstract: A system retains a connector connected to a device. The system includes a bail lock capable of connecting to the device, and a retainer capable of connecting to the bail lock. Rotation of the retainer moves the bail lock along the retainer until the connector securely attaches to the device.

Abstract: Methods and devices for processing packets are provided. The processing device may Include an input interface for receiving data units containing header information of respective packets; a first module configurable to perform packet filtering based on the received data units; a second module configurable to perform traffic analysis based on the received data units; a third module configurable to perform load balancing based on the received data units; and a fourth module configurable to perform route lookups based on the received data units.

Abstract: A system selectively drops data from a queue. The system includes queues that temporarily store data, a dequeue engine that dequeues data from the queues, and a drop engine that operates independently from the dequeue engine. The drop engine selects one of the queues to examine, determines whether to drop data from a head of the examined queue, and marks the data based on a result of the determination.

Abstract: A cable modem termination system (CMTS) provides for ordered switchover of the cable modems it serves when a primary module of the CMTS fails. When the primary module in the CMTS fails, a backup module takes over service of the cable modems. The backup module restores the service flows of the cable modems based on a restoration order determined by session recovery priorities that are related to quality of service levels required by the service flows. In particular, service flows that are engaged in VoIP “911” calls are restored first. Service flows engaged in other VoIP calls are next restored. Lastly, service flows engaged in regular data sessions are restored.

Abstract: An electronic device includes a bay for a removable component with a vertical axis of insertion and removal. The component is inserted upwardly in the electronic device with aid of an actuation mechanism. For example, the actuation mechanism may include a lever and a horizontal support member sized to hold the bottom side of the component. Rotation of the lever translates the support member in the vertical direction to smoothly lift the component into a seated position in which connectors in the component and the electronic device are coupled. The actuation mechanism may include a latch to hold the component securely in the seated position. Rotation of the lever in the opposite direction lowers the support member. The actuation mechanism may include one or more tabs on the horizontal support member or elsewhere that pull the component during removal to overcome the unmating force of the connectors.

Abstract: A system for transmitting a clock signal through a packet-based network is disclosed. The system comprises a first node configured to measure a clock frequency of the clock signal and calculate an accuracy indicator of the measured clock frequency; a second node configured to receive the clock frequency measurement and the accuracy indicator of the clock frequency measurement, and synthesize the clock signal therefrom; and a packet-based network for transmitting the measured clock frequency and accuracy indicator from the first node to the second node. A method of deriving a clock frequency by identifying packets with the shortest total transmission time is also disclosed.

Abstract: In general, techniques are described for hardware-based detection and automatic restoration of a computing device from a compromised state. Moreover, the techniques provide for automatic, hardware-based restoration of selective software components from a trusted repository. The hardware-based detection and automatic restoration techniques may be integrated within a boot sequence of a computing device so as to efficiently and cleanly replace only any infected software component.

Abstract: A connection between network nodes in a communication network is backed up. A failover label-switched path (LSP) is created starting at a first network node of a connection and ending at the second node of the connection, while bypassing the protected connection. In the event of connection failure, data is transmitted through the failover LSP. A network operator can selectively protect different types of data by using filters that bind one or more types of traffic received over one or more interfaces to different failover LSPs.

Abstract: A method may include counting the number of times each of a plurality of entries in a content addressable memory (CAM) matches one or more searches; grouping entries in the CAM into a first subset and a second subset based on the number of times each of the plurality of entries in the CAM matches one or more searches; and searching the first subset for a matching entry and, if no matching entry is found, searching the second subset for the matching entry.

Abstract: A processing engine for processing header data includes a level 2 (L2) header generation unit and a level 3 (L3) header generation unit. The L2 and L3 header generation units are implemented in parallel with one another. The L2 generation unit writes L2 header information to a first buffer and the L3 generation unit writes L3 header information to a second buffer. When the L2 and L3 header generation units finish processing a data unit, the data unit may be unloaded from the first and second buffer while a new data unit is simultaneously loaded to the header processing engine.

Abstract: A first network client requests initiation of a data transfer with a second network client. An admission control facility (ACF) responds to the initiation request by performing admission analysis to determine whether to initiate the data transfer. The ACF sends one or more packets to the second network client. In response, the second network client sends acknowledgment packets back to the ACF. The ACF performs admission analysis based on the packets sent and the acknowledgment packets, and determines whether the data transfer should be initiated based on the analysis. The admission analysis may be based on a variety of factors, such as the average time to receive an acknowledgment for each packet, the variance of the time to receive an acknowledgment for each packet, a combination of these factors, or a combination of these and other factors.

Abstract: A method may include receiving a packet associated with a flow of packets, the packet including a destination address; selecting one of a plurality of memory banks, the selected memory bank being associated with the flow of packets, wherein each of the plurality of memory banks stores the same next-hop information for forwarding the packet to the destination address; accessing, in the selected memory bank, the next-hop information for forwarding the packet to the destination address; and forwarding the packet to the destination address based on the next-hop information.

Abstract: A scheduler and method for scheduling packet forwarding operations is provided. Packet forwarding request information associated with a first set of input port/output port combinations is received. Packet forwarding request information associated with a second set of input port/output port combinations different from the first set of input port/output port combinations is received, where the first set of input port/output port combinations and the second set of input port/output port combinations are selected to not conflict with each other. Packet forwarding for both the first set of input port/output port combinations at a first future time slot and the second set of input port/output port combinations at a second future time slot are simultaneously scheduling at a first scheduler and a second scheduler, respectively, based on the received packet forwarding request information.

Abstract: A network acceleration device includes a persistent, in-memory cache of network content. For example, the cache may store content in a manner that allows a software process to map virtual memory to specific, known regions of an underlying physical memory. Upon detecting a failure of a process executing within the network device, the network acceleration device may restart the software process and remap data structures of the cache to the known regions of the physical memory without necessarily requiring that the cache content be reloaded from a non-volatile memory, such as a hard drive. In this manner, the network acceleration device may accelerate download speeds by avoiding timely cache content restoration in the event of a software process failure.

Abstract: Link failure messages are sent through a network to accelerate convergence of routing information after a network fault. The link failure messages reduce the oscillations in routing information stored by routers, which otherwise can cause significant problems, including intermittent loss of network connectivity as well as increased packet loss and latency. For example, the link failure messages reduce the time that a network using a path vector routing protocol, such as the Border Gateway Protocol (BGP), takes to converge to a stable state. More particularly, upon detecting a network fault, a router generates link failure information to identify the specific link that has failed. In some types of systems, the router communicates the link failure information to neighboring routers as well as a conventional update message withdrawing any unavailable routes. Once other routers receive the link failure information, the routers do not attempt to use routes that include the failed link.

Abstract: A multicast-capable firewall allows firewall security policies to be applied to multicast traffic. The multicast-capable firewall may be integrated within a routing device, thus allowing a single device to provide both routing functionality, including multicast support, as well as firewall services. The routing device provides a user interface by which a user specifies one or more zones to be recognized by the integrated firewall when applying stateful firewall services to multicast packets. The user interface supports a syntax that allows the user to define subsets of the plurality of interfaces associated with the zones, and define a single multicast policy to be applied to multicast sessions associated with a multicast group. The multicast policy identifies common services to be applied pre-replication, and exceptions specifying additional services to be applied post-replication to copies of the multicast packets for the one or more zones.

Abstract: Methods and apparatus for allowing routers in an autonomous system to implement LDP and RSVP at the same time. RSVP can be used in the network core with LDP being used in network regions surrounding the core. LDP LSPs are tunneled through the RSVP network core using RSVP LSPs and label stacking techniques. During route selection LDP LSPs which use an RSVP LSP tunnel are preferred over alternative LDP LSPs having an equal cost associated with them to create a preference for traffic engineered routes.

Abstract: Samples from an addressed data forwarding devices, such as a router, are associated with path-centric information. Information from the samples is used to update corresponding path-centric traffic information, such as flow information for example. The aggregated path-centric traffic information can then be used by traffic analysis operations.