International Business Machines Corporation

Kenneth Bruce MacKenzie

Richard Edwin Harper

Gregory Brian Pruett

Richard James McCarty

Amal Ahmed Shaheen

Sanaa F Abdelhadi

Ofer Biran

Ben-Ami Yassour

Guoyou Chen

Srirama Mandyam Krishnakumar

David W Nip

A computer implemented method to handle events based on historic actions. An event-handler receives an event, wherein the event has an associated event-type. The event-handler records the event to an event-action history table. The event-handler determines a history associated with the event by reading the event-action history table. The event-handler looks up a row entry in an action table based at least in part on the event-type, wherein the row entry comprises an action. The event-handler executes at least one instruction of the action. The event-handler records the event and action data to the event-action history table.

Event-handler for selecting responsive actions

Power and redundancy management policies are applied individually to the tiers of redundant servers of an application service such that power is reduced while maintaining a high level of system availability. Servers which are determined to be relatively inactive are moved to a free pool. Certain servers of the free pool are maintained in a hot standby state, while others are powered-off or set to operate in a low power mode. During times of high load, the servers in the hot standby state can be provisioned quickly into the application service.

Automated power control policies based on application-specific redundancy characteristics

A method in a computer for handling a network partition of a clustered computer system, wherein the computer is part of a number of computers located within the cluster. Data is periodically written into a data structure in a shared resource, wherein the data provides a status of the computer and wherein a data structure is associated with each cluster node. Monitoring for a partition of the cluster is performed. All data structures located in the shared resource are read in response to detecting a network partition of the cluster. A determination is made as to whether the computer is in the preferred partition, one containing the largest number of computers or is otherwise determined to be viable for continued operation. Should the computer determine that it is not a member of a preferred or otherwise viable partition, it must relinquish access to shared cluster resources requiring mutually exclusive access, such as a database on a shared disk volume. This is generally effected by the computer ceasing to operate as an active member of the cluster.

Method and apparatus for partition resolution in clustered computer systems

Growing community of inventors

Austin, TX, United States of America

Kenneth Bruce MacKenzie