International Business Machines Corporation

Michael James McDermott

Dennis R Martin

Robert Dennis Miller

Kiswanto Thayib

Michael John Snyder

Michael Steven Faunce

Glen Warren Nelson

Kevin Curtis Griffin

Robert H Satin

Thomas Richard Crowley

Robert L Wenger

Stuart D Plumlee

Bruce Marshall Walk

Michael Joseph Corrigan

Timothy John Mullins

Amartey S Pearson

Scott Dennis Helt

Duc H Nguyen

James Gregory Ranweiler

Patrick James Christenson

Russell Paul VanDuine

Mark Philip Piazza

Clinton Gene Laschkewitsch

Kenneth James Lawrence

Sherry Lynn Gordon

A computing system group member, signal bearing medium, and method. A configuration of a computing system group member is determined to be able to assume a role of a primary member of the computing system group and, in response thereto, a first status is assigned in the member. All members of the computing system group receive, in order, all messages directed to group members. A second status is assigned in response to determining that the configuration of the member is not able to support performance as a primary member. The member then processes, within the member, all messages that are transmitted to all members of the group when the member is assigned the first status and when the member is assigned the second status. The configuration of the member, while assigned either the first status or the second status, is adjusted based upon contents of the messages.

Ineligible group member status

A method backs up computer data by organizing update requests into groups. All update requests in one group complete concurrently before requests of a subsequent group complete.

Asynchronous hybrid mirroring system

A computing node that functions as a member within a computing system group, such as a cluster, that has a status allowing receipt of group messages even though the node is not an active member of the cluster. The node is able to function as a primary member or as a backup member that controls redundant resources to be utilized in case of a failure. The computing node is able to have one of two status values, an 'Active' status and an 'Ineligible' status. Members that are able to function as a primary member have an &#x201c;Active&#x201d; status assigned, and a member that is not configured or otherwise eligible to perform as a primary member is assigned an &#x201c;Ineligible&#x201d; status. Members with an Ineligible status receive all group messages and therefore are able to become configured and eligible to become a primary member.

A computing system group processing architecture that facilitates asymmetric processing at different computing nodes within a group or cluster of nodes. Nodes within a group are assigned to subgroups. Each node in a subgroup performs similar processing, but nodes in different subgroups are able to perform different processing for the same group level protocol. All nodes monitor processing completion votes that are cast by all nodes, and node in subgroups that finish processing before other subgroups synchronize to the processing of those other subgroups by casting dummy votes during vote rounds of other subgroups that are still processing their subgroup protocol.

Group synchronization by subgroups

Failover processing that accommodates failures of backup computing nodes and resources, such as data storage units and printers. Failure of a computing node that controls resources causes another computing node to assume control of the resources controlled by the failed node. Failure of the primary computing node causes another computing node, at either the same or at a different site, to be selected as the new primary node. Failure of a resource at the primary site causes the site with the next highest priority backup resource to become the new primary site. Failure of a backup computing node causes a new backup node at the same site as the failed backup node to replace the failed backup node as host for the site's resources. Backup mirroring data flows are then adjusted to reflect the new functions of the affected nodes.

Primary-backup group with backup resources failover handler

An apparatus, clustered computer system, program product and method utilize a unique prepare operation in connection with a resource action to effectively 'lock out' missing or inactive cluster entities such as nodes and cluster objects from rejoining a clustered computer system subsequent to the resource action. The prepare operation includes the modification of one or more cluster configuration parameters associated with a plurality of entities in a clustered computer system, such that any such cluster entity that is active during the prepare operation accepts the modifications, while any such cluster entity that is inactive during the prepare operation does not accept the modifications. By modifying cluster configuration parameters for active cluster entities, attempts by previously-inactive cluster entities to activate or rejoin clustering subsequent to resource actions will generally fail due to an incorrect or stale cluster configuration parameters for such entities, and as a result, such entities will be effectively blocked from being accepted into the clustered computer system.

Cluster resource action in clustered computer system incorporation prepare operation

Method and apparatus for ensuring availability of disk units in a disk pool which may be switched between a primary system and one or more backup systems. One embodiment provides a method for ensuring accessibility of one or more disk units by a system, comprising: configuring a storage pool for the system; validating availability of the one or more disk units for the storage pool; and selecting one or more valid disk units for the storage pool. The method may further comprise ranking availability of each disk unit for the storage pool and selecting one or more valid disk units for the storage pool according to availability ranking.

Method and apparatus for validating and ranking disk units for switching

A method for recovering a computer system from a loadsource direct access storage device located at a remote location. In accordance with the method and system of the present invention, after a catastrophic damage and the installation of a local loadsource direct access storage device, an indicator within the local loadsource direct access storage device is set such that the local loadsource direct access storage device cannot be utilized for initial program load. Thereafter, the contents of the entire loadsource direct access storage device located at a remote location are copied to the replacement local loadsource direct access storage device. The indicator within the replacement local loadsource direct access storage device is reset such that the replacement local loadsource direct access storage device can be utilized for initial program load again. Finally, an initial program load is performed from the replacement local loadsource direct access storage device.

Method and system for recovering a computer system from a loadsource located at a remote location

Method and apparatus for managing data on DASD units to improve system performance comprises monitoring portions of data on a plurality of DASD units to determine the times the data is accessed within a given time period, and characterizing accessed data portions of a DASD unit as HOT and COLD data. The DASD units are monitored to determine the number of times each unit is accessed within a time period to develop utilization factors reflective of the number of times the DASD unit is accessed during the time period. HOT and COLD data is moved between DASD units based on the utilization factors of the DASD units.

Dynamic DASD data management and partitioning based on access frequency utilization and capacity

DASD failures occurring on sector boundaries currently go undetected. While this sort of problem can occur at any time, it usually occurs when the program executing on the computer system's processor is operating on a multisector piece of data and periodically updating that information in auxiliary storage so that the copy in auxiliary storage is relatively up to date with the changes being made by the program. The storage management mechanism of the present invention solves the problem of sector boundary write failures by associating a sequence number with each piece of multisector data. Essentially, the sequence number becomes a property of the particular piece of data itself. When the mechanism retrieves the information from disk, it checks to make sure that all the sequence numbers match. If they do not, a data integrity problem has been detected.

Storage management mechanism that detects write failures that occur on

A method for updating mass storage device configuration records during a configuration change within a computer system is disclosed. A unit configuration record is stored in a system mass storage device. The unit configuration record is unique to each system and it includes a configuration identification, a sequence number and an alteration count table. In addition, a logical device identifier is stored in each mass storage unit within the system. The logical device identifier is unique to each mass storage device unit and it includes a configuration identification, a unit number and an alteration count. If a configuration update is necessary, the corresponding alteration count of the unit configuration record is first incremented, a new logical device identifier is then built with the new alteration count, and finally the new logical device identifier is written to a mass storage device that requires update. After the completion of configuration update, the sequence number is incremented, the updated information is then merged with the functioning copy of unit configuration record and the merged copy is subsequently written on to the system mass storage device. This invention allows a logical configuration change to be made even when the mass storage device of a configured set is unavailable, and thereby eliminating the restriction of all configured mass storage devices must be operational in order to perform a logical configuration change as required by the prior arts.

Method and system for mass storage device configuration management

A data processing system having a pair of mirrored storage units maintains a state record of the mirrored pair in system memory. In order to be able to determine state when the system is re-initialized, this state information is also stored on each storage unit of the mirrored pair, and in an alternate location. When the state changes, the operating system writes the new state to those storage units which are still functioning, and to the alternate location. In order to prevent ambiguous situations, only certain defined state transitions are permitted. When the system is re-initialized, it attempts to read the state information stored on the storage units. If either unit can not be read, the system substitutes the state retrieved from the alternate state record for the state that would have been read from the non-responding unit. This pair of states from the two units index an unique entry in a state derivation table containing the resultant state.

Method and apparatus for deriving mirrored unit state when

A method of pairing disk units in a computer system where mirroring is desirable is disclosed. Where possible, disk units are paired with other disk units located on a different bus. This provides the highest level of protection against inaccessible data due to data loss or failure of a component in the computer system. Where this is not possible, the remaining disk units are paired with disk units located on a different I/O processor, a different controller, or a different disk enclosure. Where none of these pairings are possible, the remaining disk units fall below the threshold level of protection and therefore cannot be paired within the segment. Any disk units that could not be paired are placed on a spare list. Several attempts are made to pair up disk units on the spare list. If all of these attempts fail, the paired disk units with the lowest level of protection are broken apart and added to the spare list, where attempts are again made to pair up the disk units on the spare list. After all units are paired, a pair of disk units is compared with other pairs of disk units to see if the level of protection can be increased by swapping mates. This comparison of pairs of disk units is repeated until the level of protection is optimized. Each time the computer system is started, a check is made to determine if a configuration change resulted in a lower level of protection. If so, the user is asked if he would like the disk units on the computer system to be re-paired to increase the level of protection.

Computer controlled optimized pairing of disk units

A method of pairing disk units in a computer system where mirroring is desirable is disclosed. Where possible, disk units are paired with other disk units located on a different bus. This provides the highest level of protection against inaccessible data due to data loss or failure of a component in the computer system. Where this is not possible, the remaining disk units are paired with disk units located on a different I/O processor, a different controller, or a different disk enclosure. Where none of these pairings are possible, the remaining disk units fall below the threshold level of protection and therefore cannot be paired within the segment. Any disk units that could not be paired are placed on a spare list. The paired disk units with the lowest level of protection are broken apart and added to the spare list, where attempts are again made to pair up the disk units on the spare list. After all units are paired, a pair of disk units is compared with other pairs of disk units to see if the level of protection can be increased by swapping mates. This comparison is repeated until the level of protection is optimized. Each time the computer system is started, a check is made to determine if a configuration change resulted in a lower level of protection. If so, the user is asked if he would like the disk units on the computer system to be re-paired to increase the level of protection.

Growing community of inventors

Oronoco, MN, United States of America

Michael James McDermott