International Business Machines Corporation

Jia Heng Zhong

Daniel Martin

Sowmya Kameswaran

Manogari Nogi Simanjuntak

Technology for routing queries in a system with a plurality of nodes (for example online analytical processing sub-systems) where each node has an associated replicated local database and a local latency value and replication velocity values. The workload balancing for incoming received queries among and between the plurality of nodes is based, at least in part, on consideration of latency values and/or replication velocity values for the various nodes. The best node to handle a given query is thereby selected and the query is routed to the selected node for response.

Distribution of workloads among online analytics processing systems

A computer-implemented method, system, and non-transitory computer program product for work assignment for parallelized database synchronization. Data changes to a source database are identified from a change log for the source database. The data changes are batched into data change chunks of data changes. An estimated processing cost for processing of each of the data change chunks to apply the data changes in the data change chunks to a target database is determined. A priority queue in which the data change chunks are identified in order of estimated processing cost is generated. Each data change chunk identified in the priority queue is assigned to a processing thread in a plurality of processing threads for processing in order by estimated processing cost from higher estimated processing cost to lower estimated processing cost to apply the data changes in the data change chunks to the target database.

Work assignment in parallelized database synchronization

A method, system, and program product for enabling multiple source database coherency is provided. The method includes detecting an incoming query to be executed with respect to an accelerator server. An initial timestamp of the incoming query is analyzed and additional timestamps of database tables referenced by the incoming query are retrieved. It is determined that each timestamp of the additional timestamps is reached with respect to a corresponding replication engine and target data from the database tables is synchronized with source data of a source database. In response, the incoming query is accelerated for execution with respect to a target database management system.

Growing community of inventors

San Jose, CA, United States of America

Jia Heng Zhong