International Business Machines Corporation

Yongli An

Mohammad Khatibi

Jesse Feng Chen

Peter Kin Leung Shum

Customer data volume of a database and estimated transaction throughput requirements for a multi-tiered transaction system are received at a server sizing computing device. The customer data volume is compared to the estimated transaction throughput requirements. Based upon the comparison of the customer data volume to the estimated transaction throughput requirements, a customer data volume edge case for which server hardware is improperly sized by transaction-based hardware sizing is detected. Using the customer data volume, a quantity of operational memory of a server sufficient to store the customer data volume generated under the detected customer data volume edge case during real-time transactional processing by the server is calculated.

Data volume-based server hardware sizing using edge case analysis

Systematically enforcing generation of partition key values for records in a data management system such that for any given partitioned table, all records relating to a common business object share a common partitioning key. Partitioning and/or clustering of these records can then be done based on these specially generated partitioning key values, with entities dependent on the same first-class entity being placed in the same partition and/or cluster. This may provide considerable performance gains in at least some circumstances.

Data partitioning for high-efficiency data retrieval

Disclosed is a queuing system implemented method, a queuing system and an article of manufacture. The queuing system implemented method directs the queuing system to manage a queue according to a quality of service parameter. The queue is associated with the queuing system. The queue is used for receiving requests. The queue is associated with a worker thread used to process contents of the queue. The queuing system implemented method includes: placing the request in the queue, permitting the worker thread to process the placed request, determining a response time, the response time being the time taken to complete processing of the placed request, and adjusting the worker thread if the determined response time does not satisfy the quality of service parameter, the adjusted worker thread used for processing another request placed in the queue in which a new response time taken to complete processing of the another placed request satisfying the quality of service parameter.

Growing community of inventors

Markham, Canada

Yongli An