International Business Machines Corporation

Steve Raspudic

Dale Murray McInnis

Steven Robert Pearson

Vincent Kulandaisamy

Markus Mueller

Yuke Zhuge

Jon A Lind

Andrew David Buckler

Jonna Maaria Judin

Fiona Kennedy

Slawomir Piotr Maludziński

Mark John Mitten

Subho Chatterjee

Anand Subramanian

According to one aspect of the present disclosure, a method and technique for clustered computer environment partition resolution is disclosed. For a clustered environment having a plurality of nodes, the method includes: creating a key by a first node of the plurality of clustered nodes, the plurality of clustered nodes each configured to have access to a storage provider resource; communicating the key to remaining nodes of the plurality of clustered nodes; responsive to detecting a potential partition event, generating by at least one of the plurality of clustered nodes a lock code using the key for reserving the storage provider resource; and responsive to determining an unlocked status of the storage provider resource, resolving the partition event by writing the lock code to the storage provider resource.

Clustered computer environment partition resolution

Described are systems and methods for communicating between a primary database and a standby database of a high availability data recovery (HADR) system. A plurality of primary partitions in a primary database and a plurality of standby partitions in a standby database are configured to communicate with each other. A transition of the plurality of primary partitions from a first HADR state to a second HADR state is synchronized.

High availability database systems and methods

Generating in a computer system and deploying a data model of a plurality of database cluster configuration availability solutions over a computer network by creating a database cluster configuration modeling specification including objects contained in a unified model language diagram providing a definition of a database cluster configuration data model. Creating the database cluster configuration data model using the database cluster configuration modeling specification and upon receiving signals from a graphical user interface or from XML batch data files or from application programming interfaces, indicating the definition of the cluster configuration model. After creating the graphical database cluster configuration data model, constraints that specify a valid logical configuration are validated and the solution is transformed into data understandable by database cluster manager software applications and then transmitted over a network to multiple cluster manager targets to be incorporated as database cluster configuration availability solutions.

Systems, methods and computer products for database cluster modeling

A method, system, and computer program product for reducing likelihood of data loss during performance of failovers in a high-availability system comprising a primary system and a standby system are provided. The method, system, and computer program product provide for defining a halt duration, periodically determining a halt end time, halting data modifications at the primary system responsive to failure of data replication to the standby system, resuming data modifications at the primary system responsive to a last determined halt end time being reached or data replication to the standby system resuming, and responsive to the primary system failing prior to a previously determined halt end time, determining that a failover to the standby system will not result in data loss on the standby system with respect to the primary system.

Reducing likelihood of data loss during failovers in high-availability systems

Growing community of inventors

Mississauga, Canada

Steve Raspudic