Abstract: In a message delivery system, wherein messages are published by publishers and delivered to subscribers based on subscriptions relating to message content. A last message queue is associated with each publisher, and a last message from each publisher is stored in the message queue associated with that publisher based on an attribute of the publisher. In this way, it is not necessary to modify the content data within the message to create the last message queue.

Abstract: In a message delivery system wherein messages having topic portion defining a topic and a body portion containing information relating to the defined topic are delivered to subscribers based on the topics associated with the messages, the messages are placed in a queue for delivery, and in the event of congestion one or more of the messages are replaced with discard indication messages containing the topics associated with the discarded messages without the corresponding body portions.

Abstract: A method of managing shared resources for controlling message flows in an assured message delivery system having a plurality of clients involves identifying different classes of clients based on their performance behaviors, and providing differentiated access to said shared resources between said different classes of clients.

Abstract: In a message delivery system, messages are published by publishers and delivered to subscribers based on subscriptions relating to message content. Activation and de-activation notifications are supplied to applications that are bound to a queue in the message delivery system.

Abstract: In a message delivery system, messages are carried over a network in packets. Server hosts support applications for originating and receiving messages and network interfaces. Switches interconnect the server hosts and form the packet network. Some of the switches are linked to the local server hosts. Distributed filtering/matching modules associated with the local server hosts register interests for applications on the associated local server hosts, inspect packets received over the network or from a local server host to match messages with registered interests and remove any unmatched packets from the packets. The matched messages are forwarded to the local server host for delivery to the applications. This system offers the ability to use special purpose hardware for the filtering/matching function without requiring connections to a common message broker.

Abstract: An assured message delivery system with resiliency receives messages from publishing client(s) and delivers them to subscribing clients. A primary system delivers the messages during normal operation. A secondary system mirroring the configuration of the primary system takes over responsibility for delivering the messages in the event of failure of the primary system. The secondary system is connected to the primary system over a communications link. The primary system sends copies of the messages to the secondary system over the communications link. The secondary system stores replicate messages and sends acknowledgement messages back to the primary system.

Abstract: In a message delivery system wherein messages having topic portion defining a topic and a body portion containing information relating to the defined topic are delivered to subscribers based on the topics associated with the messages, the messages are placed in a queue for delivery, and in the event of congestion one or more of the messages are replaced with discard indication messages containing the topics associated with the discarded messages without the corresponding body portions.

Abstract: A method of assigning a priority to a document in a content-routed network including a plurality of routers interconnected by links, involves matching rules against the content of a received document to assign a priority, mapping the priority onto an egress link, and mapping the egress links onto the interconnect network between content routers to achieve end-to-end quality of service.

Abstract: A method of propagating multicast subscription and routing information between routers in a network, and constructing forwarding tables in the routers, allowing providers of the data (the publishers), and the multicast recipients of the data (the subscribers) to be decoupled from and have no knowledge of one another. This is done without the need to maintain (Source Network, Published-Multicast-Group) State in the routers, for a highly scalable solution in those applications where there is a possibility for a large, or infinite number of Published-Multicast-Groups, and in those applications where messages are being routed by content, so it is impossible to identify published-multicast-groups.

Abstract: In method of providing assured message delivery across a distributed message delivery system with low delivery latency and network traffic, a set of destinations is first identified for a set of destinations for a message received at an ingress router to the network. The received message is stored in persistent storage along with meta-data about each destination for the message before the message is routed to each identified destination. The message is only removed from persistent storage when an acknowledgement has been received from each destination indicating that the message has been successfully received.

Abstract: A networking appliance includes a chassis, an off-the-shelf motherboard mounted on said chassis, said motherboard including at least one expansion bus, one or more separately removable expansion module mounted on said chassis, and a number of card slots in the expansion module for accommodating expansion cards. A connection arrangement provides a signal connection between the card slots in the expansion modules and the expansion buses.

Abstract: A method and apparatus for converting documents from one format to another in a speed efficient way involves a hardware module which implements several operating pipeline stages which work in parallel. The transformations are supplied and decomposed into sequences of control units. The transformation of documents consists of applying control unit sequences to input documents. The control units are themselves executed by a set of dedicated hardware resources. Furthermore the pipeline is capable of operating on more than one document at a time. Fast document transformation is a key capability of document processing systems. The use of parallel processing techniques and hardware that implements highly specialized transformation resources make this invention particularly scalable for its use in large, high speed content networks.

Abstract: A method of performing connectivity verification in a content routed network including a plurality of routers and wherein messages are transported through the network over a message transport layer, involves sending from an originating router a ping request message over the message transport layer along a user data path to a terminating router using a content routing protocol. The terminating router responds with a ping reply message upon receipt of the ping request message, and the message transport layer is monitored for the ping reply message at the originating router. Trace messages can also be sent through the network to determine whether subscribers are reachable for messages originating from a particular publisher. Also trace messages can be sent through the network to determine whether subscriptions from a particular subscriber are reachable from a particular originating router.

Abstract: A method of managing a content routed network, involves distributing published documents through said network for delivery to subscribers; maintaining data logs pertaining to said published documents at different points in the network; and correlating the data logs to obtain information about the operation of the network.

Abstract: In a method of managing a content-based network including a plurality of content routers, such as XML routers, dynamic links are established between the content routers which are not direct neighbors.

Abstract: A method of providing a shared application infrastructure simultaneously to multiple independent virtual private networks with potentially overlapping IP addresses, involves connecting to an underlying layer 1, 2 or 3 virtual private network offering one or more virtual private networks, determining which virtual private network traffic is being received from or sent into, maintaining logically separate routing tables per VPN to allow overlapping addresses, and providing an application which provides a logical instance of the service to each VPN.

Abstract: A method of providing router redundancy within a distributed network of routers, wherein messages are routed within the network based on virtual router identifiers, involves organizing the routers into one or more redundancy groups; assigning a physical identifier to each router in each of the redundancy groups; assigning one or more virtual identifiers to each redundancy group; selecting one of the routers of a particular redundancy group as a currently active router associated with a particular virtual identifier assigned to that redundancy group; advertising among the distributed network of routers for each redundancy group the physical identifier for the active router and information enabling other routers to determine the virtual identifier with which the currently active router is associated; and forwarding messages destined for a particular virtual router identifier to the currently active router based on the physical identifier of the currently active router.

Abstract: A carrier grade content router, includes a distinct management plane for housing externally visible management applications, and coordinating and relaying external management requests to appropriate underlying application code in the router; a distinct control plane for running control protocols required within an XML routed network; and a distinct data plane for receiving and forwarding customer data. Some functions can be implemented in software or via a hardware accelerator.

Abstract: In a method of managing a content-based network, which is typically XML-based, and optionally may be overlaid on an underlying network having a plurality of network elements interconnected by links, a link state protocol maintains each network element's topological view of the overlay network from the underlying network. A subscription management protocol ensures dissemination of published content within the content-based network independently of the link state protocol.

Abstract: In a method of performing the asynchronous delivery of messages from a server to a client, the server receives the messages for the client from a message source. The server and client communicate over a packet network using an asymmetric protocol that requires request messages to be originated from the client to the server. At least one request message is sent in advance from the client to the server, and these request messages are stored at the server. When the server receives a message for the client from the message source, it forwards the message to the client as a response to a stored request message.