Abstract: In a communication system having a plurality of physical communication ports, a hierarchical bond communication interface has a logical interface as a component of the bond at a top level of the hierarchy, and a first subjugate logical interface at a second level of the hierarchy as a component of the top-level logical interface. The communication system has a control mechanism whereby thresholds may be set to determine up or down status of hierarchical bonds for use in communication, according to characteristics of components of the bonds.

Abstract: A method for managing data traffic in nodes in a fabric network, each node having internally-coupled ports, follows the steps of establishing a managed queuing system comprising one or more queues associated with each port, for managing incoming data traffic; and accepting or discarding data directed to a queue according to the quantity of data in the queue relative to queue capacity. In one preferred embodiment the managed system accepts all data directed to a queue less than full, and discards all data directed to a queue that is full. In some alternative embodiments the queue manager monitors quantity of data in a queue relative to queue capacity, and begins to discard data at a predetermined rate when the quantity of queued data reaches the threshold. In other cases the queue manager increases the rate of discarding as the quantity of queued data increases above the preset threshold, discarding all data traffic when the queue is full.

Abstract: In a communication system having a plurality of physical communication ports, a hierarchical bond communication interface has a logical interface as a component of the bond at a top level of the hierarchy, and a first subjugate logical interface at a second level of the hierarchy as a component of the top-level logical interface. The communication system has a control mechanism whereby thresholds may be set to determine up or down status of hierarchical bonds for use in communication, according to characteristics of components of the bonds.

Abstract: In a communication system having a plurality of physical communication ports, a hierarchical bond communication interface has a logical interface as a component of the bond at a top level of the hierarchy, and a first subjugate logical interface at a second level of the hierarchy as a component of the top-level logical interface. The communication system has a control mechanism whereby thresholds may be set to determine up or down status of hierarchical bonds for use in communication, according to characteristics of components of the bonds.

Abstract: An inter-network method for distributing a path-vector routing protocol within a multi-processor router has steps of running an instance of a router-distributed Border Gateway Protocol (rdBGP) on each of a plurality of processors of the multi-processor router, establishing routing sessions between individual instances of rdBGP running on separate processors of the multi-processor router, and between individual ones of the instances of rdBGP and external protocol peers, exchanging routing information between instances of rdBGP and the external protocol peers, wherein the instances of rdBGP learn, from the external protocol peers, routes for a plurality of destinations, calculating independently, by each instance of rdBGP, a set of routes, including a best route, from each of the plurality of destinations learned from the external protocol peers, and sharing route information determined with the other instances of rdBGP running on separate processors in the multi-processor router, and updating route information

Abstract: A control system is disclosed for controlling data flow over a data-packet-network according to specific destinations. The system has a network monitoring system for monitoring network performance parameters; a network access system for accessing specific nodes in the network; and a control software executable on the network access system for assigning and changing cost parameters at selected nodes in the network. A network administrator uses the network access system and control software to assign and implement cost values at the selected nodes, the values associated individually with a specific destination or destinations, the values establishing forwarding costs to be incurred at the selected nodes, and link costs to be incurred per data link between the nodes such that manipulation of such cost value assignments enables load balancing of data traveling through the network. In one embodiment the data-packet-network is internal to a router and nodes comprise internal router cards.

Abstract: A method of and apparatus for distributing data for a database between a plurality of processors in a distributed processor system involves running a database management system on a first plurality of processors in conjunction with a plurality of protocols that generate or amend data for the database. Data is distributed from servers to clients registered in a server-client relationship. Server-client relationships may also be registered between a second, and a third plurality of processors that do not run protocols generating or amending data for the database. Fault tolerant redundancy is provided by clients registering with two or more servers, one as a primary and another as a secondary, and activating the secondary if the primary fails. The method is particularly applicable to scalable data packet routers having a plurality of processors operating on different line and control cards.

Abstract: A data structure has a list of event objects, one or more producers creating the event objects for the list, and a finite set of consumers accessing the object list. The structure is characterized in that the event objects are each associated with a first reference number indicating the number of consumers currently accessing the event object, and in that, as each consumer completes access the reference number is decremented, and when the reference number for an event object is zero, and the event object is at the head of the list, indicating that all consumers have accessed the object, that event object is removed from the list. A use of the structure for propagating route changes to nodes in a multiple-processor router, and to peer routers in a network is also taught.

Abstract: An inter-network method for distributing a path-vector routing protocol within a multi-processor router has steps of running an instance of a router-distributed Border Gateway Protocol (rdBGP) on each of a plurality of processors of the multi-processor router, establishing routing sessions between individual instances of rdBGP running on separate processors of the multi-processor router, and between individual ones of the instances of rdBGP and external protocol peers, exchanging routing information between instances of rdBGP and the external protocol peers, wherein the instances of rdBGP learn, from the external protocol peers, routes for a plurality of destinations, calculating independently, by each instance of rdBGP, a set of routes, including a best route, from each of the plurality of destinations learned from the external protocol peers, and sharing route information determined with the other instances of rdBGP running on separate processors in the multi-processor router, and updating route information

Abstract: A method for managing data traffic in nodes in a fabric network, each node having internally-coupled ports, follows the steps of establishing a managed queuing system comprising one or more queues associated with each port, for managing incoming data traffic; and accepting or discarding data directed to a queue according to the quantity of data in the queue relative to queue capacity. In one preferred embodiment the managed system accepts all data directed to a queue less than full, and discards all data directed to a queue that is full. In some alternative embodiments the queue manager monitors quantity of data in a queue relative to queue capacity, and begins to discard data at a predetermined rate when the quantity of queued data reaches the threshold. In other cases the queue manager increases the rate of discarding as the quantity of queued data increases above the preset threshold, discarding all data traffic when the queue is full.

Abstract: A method for managing data traffic in nodes in a fabric network, each node having internally-coupled ports, follows the steps of establishing a managed queuing system comprising one or more queues associated with each port, for managing incoming data traffic; and accepting or discarding data directed to a queue according to the quantity of data in the queue relative to queue capacity. In one preferred embodiment the managed system accepts all data directed to a queue less than full, and discards all data directed to a queue that is full. In some alternative embodiments the queue manager monitors quantity of data in a queue relative to queue capacity, and begins to discard data at a predetermined rate when the quantity of queued data reaches the threshold. In other cases the queue manager increases the rate of discarding as the quantity of queued data increases above the preset threshold, discarding all data traffic when the queue is full.

Abstract: An inter-network method for distributing a path-vector routing protocol within a multi-processor router has steps of running an instance of a router-distributed Border Gateway Protocol (rdBGP) on each of a plurality of processors of the multi-processor router, establishing routing sessions between individual instances of rdBGP running on separate processors of the multi-processor router, and between individual ones of the instances of rdBGP and external protocol peers, exchanging routing information between instances of rdBGP and the external protocol peers, wherein the instances of rdBGP learn, from the external protocol peers, routes for a plurality of destinations, calculating independently, by each instance of rdBGP, a set of routes, including a best route, from each of the plurality of destinations learned from the external protocol peers, and sharing route information determined with the other instances of rdBGP running on separate processors in the multi-processor router, and updating route information

Abstract: A control system is disclosed for controlling data flow over a data-packet-network according to specific destinations. The system has a network monitoring system for monitoring network performance parameters; a network access system for accessing specific nodes in the network; and a control software executable on the network access system for assigning and changing cost parameters at selected nodes in the network. A network administrator uses the network access system and control software to assign and implement cost values at the selected nodes, the values associated individually with a specific destination or destinations, the values establishing forwarding costs to be incurred at the selected nodes, and link costs to be incurred per data link between the nodes such that manipulation of such cost value assignments enables load balancing of data traveling through the network. In one embodiment the data-packet-network is internal to a router and nodes comprise internal router cards.

Abstract: A data structure has a list of event objects, one or more producers creating the event objects for the list, and a finite set of consumers accessing the object list. The structure is characterized in that the event objects are each associated with a first reference number indicating the number of consumers currently accessing the event object, and in that, as each consumer completes access the reference number is decremented, and when the reference number for an event object is zero, and the event object is at the head of the list, indicating that all consumers have accessed the object, that event object is removed from the list. A use of the structure for propagating route changes to nodes in a multiple-processor router, and to peer routers in a network is also taught.

Abstract: A method for managing data traffic in nodes in a fabric network, each node having internally-coupled ports, follows the steps of establishing a managed queuing system comprising one or more queues associated with each port, for managing incoming data traffic; and accepting or discarding data directed to a queue according to the quantity of data in the queue relative to queue capacity. In one preferred embodiment the managed system accepts all data directed to a queue less than full, and discards all data directed to a queue that is full. In some alternative embodiments the queue manager monitors quantity of data in a queue relative to queue capacity, and begins to discard data at a predetermined rate when the quantity of queued data reaches the threshold. In other cases the queue manager increases the rate of discarding as the quantity of queued data increases above the preset threshold, discarding all data traffic when the queue is full.

Abstract: In a communication system having a plurality of physical communication ports, a hierarchical bond communication interface has a logical interface as a component of the bond at a top level of the hierarchy, and a first subjugate logical interface at a second level of the hierarchy as a component of the top-level logical interface. The communication system has a control mechanism whereby thresholds may be set to determine up or down status of hierarchical bonds for use in communication, according to characteristics of components of the bonds.