International Business Machines Corporation

Kevan Lee Miller

Michael James Ward

Robert Evan Strom

Guruduth Somasekhara Banavar

Daniel Charles Sturman

Tushar Deepak Chandra

Atul Prakash

Josyula R Rao

Donald Philip Pazel

Michael S Meier

James Robert Russell

A publish/subscribe system is non-disruptively reconfigured, such that messages are not lost during the reconfiguration. After reconfiguration is initiated, messages that are to be delivered within the publish/subscribe system are forwarded using either an old path or a new path. That is, until the reconfigured system is stabilized, some messages may be delivered using the old path, while other messages are delivered using a new path. After the system is stabilized, the new path is used for delivery of all the messages, until another reconfiguration is initiated.

Non-disruptive reconfiguration of a publish/subscribe system

A technique for reliably multicasting a message within a router network is provided. At least one special router in the network has associated persistent storage for logging a message being routed to one or more clients. When a message is received at this logger node, the logger places the message in persistent storage and sends a logging acknowledgment back to those routers to which the message was originally routed, as well as back towards the source of the message. The logger acknowledgment includes the message id, the logger id, and a logging number. When the logger acknowledgment is received by a router, it looks up the routing information from the original message, and sends the acknowledgment to those neighboring routers to which the original message was sent, excluding the link from which the logger acknowledgment was received. If the original message corresponding to the logger acknowledgment was buffered at this router waiting to be delivered, the message is then delivered to its client nodes. Processes for recovering from node or link failure within the router network, and for at most once delivery of messages, at least once delivery of messages, and exactly once delivery of messages are also provided.

Message logging for reliable multicasting across a routing network

A system and method featuring a collection of model objects which hold data to be shared, and supporting interfaces to read and write the data, to create and destroy collaboration relationships and to attach 'view objects' containing application code which will be automatically notified of value changes. Each user environment is provided with a model object control entity, referred to herein as a Collaboration Manager, which creates model objects, executes transactions, and creates association objects related to the collaboration. A concurrency control mechanism records the reads and updates of the data and of collaboration relationships within a single transaction site and guarantees that they are applied atomically and consistently to both the local and to all remote replica objects of the collaboration.

System and method for providing collaborative replicated objects for synchronous distributed groupware application

An adaptive view mechanism can dynamically adapt to changing conditions and provide the benefits of both optimistic and pessimistic notifications. This can, for example, provide a consistent and jitter-free GUI to display the state of distributed replicated data objects in the presence of concurrency control conflicts and losses. In a Model-View-Controller architecture of building distributed collaborative applications where model objects are replicated at distributed sites, adaptive views can dynamically switch between behaving like pessimistic views and like optimistic views. When the number of concurrent updates is low, the adaptive views can be made to operate in an optimistic mode&#x2014;providing good responsiveness to users. However, during those periods of heavy activity, when the number of concurrent updates becomes high, the adaptive view can be made to operate in a pessimistic manner&#x2014;providing more reliable response to users. For each adaptive view, a programmer can specify a policy object that can monitor the prevailing conflict rate and loss rate as if the adaptive view were behaving optimistically. Using these rates, the policy object can apply arbitrary application logic or thresholds to determine whether it is beneficial for the adaptive view to behave optimistically or pessimistically. The policy object can then request the adaptive view to dynamically adapt its behavior to the appropriate manner. In general however, the dynamic switching can be based on any application specific request to switch modes. Distributed sites that have replicated copies of the same model object can view the replica heterogeneously, that is, pessimistically, optimistically, or adaptively, according to the user's requirements.

Dynamically adapting between pessimistic and optimistic notifications to replicated objects

Reconfiguration of a continuously available routing network of a publish/subscribe system is achieved by quiescing data messages and control messages within the routing network to preserve first-in first-out ordering thereof, and then, reconfiguring the quiesced routing network. This reconfiguration technique is particularly applicable to a routing network having at least one special node comprising one of a logger node or a sequencer node, wherein the logger node is used to store a message to persistent storage within the routing network prior to delivery of the message to one or more clients, and the sequencer sequences a message with other messages received by the network prior to delivery thereof for ordered delivery of the messages to one or more clients of the network.

Quiescent reconfiguration of a routing network

A technique for reliably multicasting a message within a router network in a sequenced order with other messages is provided. At least one special router in the network has associated persistent storage for sequencing a message being routed to one or more clients. When a message is received at this sequencer node, the sequencer places a sequence number for the message in persistent storage and sends a sequencing acknowledgment back to those routers to which the message was originally routed, as well as back towards the source of the message. The sequencer acknowledgment includes the message id, and the sequencing number. When the sequencer acknowledgment is received by a router, it looks up the routing information from the original message, and sends the acknowledgment to those neighboring routers to which the original message was sent, excluding the link from which the sequencer acknowledgment was received. If the original message corresponding to the sequencer acknowledgment was buffered at this router waiting to be delivered, the message is then delivered to its client nodes.

Message sequencing for ordered multicasting of a message across a routing network

A system and method featuring a collection of model objects which hold data to be shared, and supporting interfaces to read and write the data, to create and destroy collaboration relationships and to attach &ldquo;view objects&rdquo; containing application code which will be automatically notified of value changes. Each user environment is provided with a model object control entity, referred to herein as a Collaboration Manager, which creates model objects, executes transactions, and creates association objects related to the collaboration. A concurrency control mechanism records the reads and updates of the data and of collaboration relationships within a single transaction site and guarantees that they are applied atomically and consistently to both the local and to all remote replica objects of the collaboration.

System and method for providing collaborative replicated objects for synchronous distributed groupware applications

This invention provides a small and efficient set of primitives for debugging a distributed application that runs on a plurality of processors connected by a common network. These primitives permit a user to debug a distributed application in a manner similar to debugging a non-distributed application. The invention allows a user to step into and return from a remote procedure call in exactly the same manner as a local procedure call. The invention also allow a user to set breakpoints in a server for specific clients and to specify conditions under which a particular remote call from a client should trap in a server. These capabilities greatly simplify the process of debugging distributed applications such as client-server programs.

Growing community of inventors

Danbury, CT, United States of America

Kevan Lee Miller