International Business Machines Corporation

Mark C Astley

Sumeer K Bhola

Robert Evan Strom

Cristian Lumezanu

Marc Adam Kaplan

Joshua S Auerbach

Daniel Charles Sturman

Yuanyuan Zhao

Tushar Deepak Chandra

Ignacio Silva-Lepe

Michael James Ward

David Ware

Andrew D J Banks

A system and method for latency assignment in a system having shared resources for performing jobs including computing a new resource price at each resource based on latencies in a previous iteration. The new resource price is sent to a task controller that has at least one subtask running at the resource. Each subtask has an initial deadline. New deadlines for the subtasks in a task are determined based on the new resource prices at the task controller. The new deadlines are sent to the resources where at least one subtask in the task is running to improve system performance.

Distributed online optimization for latency assignment and slicing

A system and method for latency assignment in a system having shared resources for performing jobs including computing a new resource price at each resource and sending the new resource price to a task controller in a task path that has at least one job running in the task path. A path price is computed for each task path of the task controller, if there is a critical time specified for the task. New deadlines are determined for the resources in a task path based on the resource price and the path price. The new deadlines are sent to the resources where the at least one job is running to improve system performance.

A system and method which integrates a greedy allocation for consumer admission control with a Lagrangian allocation to compute flow rates and which links the results of the two approaches to allow a tradeoff between consumer admission control and flow rates. The Lagrangian Rates, Greedy Populations (hereinafter, 'LRGP') solution is a scalable and efficient distributed solution for maximizing the total utility in an event-driven distributed infrastructure. The greedy population, consumer portion generates prices used in the LaGrangian rate flow approach. The method is iterative including a regular exchange of information for ongoing optimization, dynamically adjusting producer rates in response to changes to consumer service and dynamically adjusting the service to consumer populations in response to changes in the producer rates.

System and method for distributed utility optimization in a messaging infrastructure

A method, system, and computer program for delivering messages and data to competing consumers. One aspect of the invention includes a message delivery system including a destination messaging engine, one or more receiver messaging engines, and a message pool. The destination messaging engine is configured to distribute data from one or more producers. The receiver messaging engines are configured to request data from the destination messaging engine and transmit the data to one or more consumers. The message pool is configured to store data from the producers, with the destination messaging engine arbitrating data in the message pool among the receiver messaging engines.

Mechanism for delivering messages to competing consumers in a point-to-point system

A content-based publish/subscribe system for providing gapless message delivery from a publishing client to a subscribing client is disclosed. The system includes a plurality of brokers operatively coupled to one another via a network. Each of the brokers include at least one processor and are configured as a publisher-hosting broker, a subscriber-hosting broker or an intermediate broker. The subscriber-hosting broker may be operatively coupled to the publisher-hosting broker via the network through the intermediate broker. At least a portion of the plurality of brokers are configured so as to eliminate a need for persistent storage of messages at the intermediate broker and to substantially guarantee a gapless delivery of one or more messages transmitted by the publishing client to the subscribing client, even in the presence of a failure in the publish/subscribe system.

Gapless delivery and durable subscriptions in a content-based publish/subscribe system

A content-based publish/subscribe system for providing gapless message delivery from a publishing client to a subscribing client, the system includes a plurality of brokers operatively coupled to one another via a network. Each of the brokers include at least one processor and are configured as a publisher-hosting broker, a subscriber-hosting broker or an intermediate broker. The subscriber-hosting broker is operatively coupled to the publisher-hosting broker via the network through the intermediate broker. At least a portion of the plurality of brokers are configured so as to eliminate a need for persistent storage of messages at the intermediate broker and to substantially guarantee a gapless delivery of one or more messages transmitted by the publishing client to the subscribing client, even in the presence of a failure in the publish/subscribe system.

A facility is provided for distributing events to consumers in a content-based publish-subscribe system, wherein the consumers each have at least one subscription. The facility includes deriving a set of g approximate multicast groups from a larger set of G possible multicast groups in the publish-subscribe system. The deriving includes exploiting knowledge of subscription predicates of the consumers of the publish-subscribe system. The set of G possible multicast groups is collapsed to the smaller set of g approximate multicast groups, while minimizing the expected performance penalty in using the approximate multicast groups. The set of g approximate multicast groups is then used to forward events to consumers within the publish-subscribe system.

Method and system for matching consumers to events employing content-based multicast routing using approximate groups

A generalized search data structure is used to match consumers to events in event computing systems. The search data structure includes one or more paths from a root of the structure to one or more leaves of the structure. Each path has at least one level and each level corresponds to a filter attribute. The value of at least one filter attribute in at least one path is a don't care value indicating traversal of that path is guaranteed to proceed. In addition to following the path with the don't care value, one or more additional paths may also be followed. Thus, traversal of the search data structure may yield zero or more results, indicating that zero or more consumers match the specified event. Various optimizations of the search data structure are possible.

Growing community of inventors

Wayne, NJ, United States of America

Mark C Astley