International Business Machines Corporation

Samuel T Massey

Martin Andrew Ross

Craig H Stirling

Daniel J McGinnes

David Granshaw

Richard Graham Schofield

A messaging system comprises a plurality of connected components and including a schema defining fields for messages, at least one field defined as non-essential. A mechanism for operating the messaging system comprises the steps of collecting one or more performance metrics for one or more components of the messaging system, determining that at least one performance metric has crossed a predetermined threshold, informing one or more components of the messaging system that a surge in workload has occurred, and the informed components removing non-essential fields from transmitted messages and/or not processing non-essential fields from received messages.

Workload management

A computer-implemented method of parsing a message comprising a sequence of data fields, the method comprising evaluating program code for processing the parsed message to identify a first set of data fields of the message that are referenced in said program code; identifying the boundaries of the data fields in a schema defining the format of said message; identifying a second set of data fields in said schema related to the first set of data fields by reference, said second set further including the first set; and sequentially parsing the message using the identified data field boundaries, wherein said parsing step comprises skipping data fields in said sequence that precede the first data field belonging to the second set. A computer program product comprising program code for implementing this method and a data processing system adapted to implement this method are also disclosed.

Opaque message parsing

Managing a system comprising a pool of servers including a number of active servers. Utilization of the system capacity is monitored, in which the utilization results from a workload of the number of active servers. Upon detecting a critical utilization of the system capacity during the monitoring of the utilization of the system capacity, a duration of the detected critical utilization of the system capacity is predicted from a monitoring history of the utilization of the system capacity. Further, the number of active servers is increased, if the predicted duration exceeds a defined threshold. Tuning measures may also be deployed to handle surges in capacity utilization.

Server pool management

Allocating threads to processors based, at least in part, on identifiers for thread sets and applications. A thread is paired with an application and, using the identifier for the application, an ID pairing is allocated to a processor.

Providing additional thread context to assist memory locality

Event processing is prioritized based on system workload. A time constraint attribute is defined in an event rule. The event rule uses one or more events. An event processing system is monitored to determine when the system is under a predefined level of stress. If the system is determined to be under the predefined level of stress, the time constraint attribute in the event rule is used to establish when the processing of a received event used in an event rule must be carried out.

Prioritising event processing based on system workload

Processing events for use with a complex event processing system that includes a server computer system and a client application executable on a client computer system. The processing includes analyzing content of the events and assigning a plurality of attributes to associated event processing logic. A first attribute is associated with event processing logic that is operable to be applied at the client computer system and a second attribute is associated with event processing logic that is operable to be applied at a server computer system. The processing also includes using, in response to receiving a first event, assigned attributes to determine whether event processing logic associated with the first event is operable to be applied at the client computer system. The first event is forwarded to the server computer system based on determining that the event processing logic associated with the first event is not operable to be applied at the client computer system.

Apparatus for processing one or more events

Embodiments relate to controlling persisting of data to disk for use with a messaging system including an application operable for generating and sending a message to a destination. An aspect is an apparatus, which includes a comparator operable to determine whether the message includes common data, and an extractor, responsive to a determination that the message includes common data, configured to extract the common data and non-common data from the message. The apparatus also includes a persistor operable to determine whether the common data have already been persisted to disk. The persistor is operable, in response to a determination that the common data have not already been persisted to disk, to separately persist the common data and the non-common data to disk. The persistor is operable, in response to a determination that the common data have already been persisted to disk, to persist the non-common data to disk.

Controlling persisting of data to disk

A computer captures a thread state data of a first program. The computer generates a second program by applying a first program patch to the first program. The computer inserts additional code into a source code of the second program. The computer executes a first performance test using the second program and the inserted additional code. The computer monitors a first status of the second program and the inserted additional code that are under the first performance test. The computer determines a degree that a first deadlock state was prevented during the first performance test. The computer generates a first message based, at least in part, on the determined degree that the first deadlock state was prevented during the first performance test.

Using application state data and additional code to resolve deadlocks

A system for workflow placement in a computer system wherein the performance of a system processing a workflow is monitored against one or more performance measures. One or more logic elements from the workflow are delegated for subsequent processing to other systems in the computer system in response to one or more of the one or more performance measures being within a predetermined range.

Growing community of inventors

Winchester, United Kingdom

Samuel T Massey