Welcome to the 2025 IDUG AU Db2 Tech Conference! Enjoy two days full of technical sessions in both Sydney and Canberra as well as complimentary workshops in Sydney that are available to all attendees between the two cities.
Sign up or log in to bookmark your favorites and sync them to your phone or calendar.
The Db2 Version 12.1 introduces an innovative AI-powered query optimizer, marking a significant leap in database performance tuning. Inspired by the precision and expertise required in diamond cutting, the AI query optimizer redefines the cardinality estimation process, a cornerstone of cost-based optimization. Leveraging artificial neural networks, this solution learns data distributions and correlations across columns, enabling more accurate predictions for a variety of predicate types, including equality, range, and complex combinations. Integrated seamlessly with traditional methods, the AI optimizer addresses historical challenges, such as statistical correlations and range predicates, without disrupting existing workflows. Its automatic training framework identifies strongly correlated columns and builds models using sample-based data, ensuring efficiency and scalability. Experimental results reveal a 3x improvement in query performance for workloads with complex predicates, demonstrating the transformative potential of AI in optimizing query execution plans. This breakthrough sets the stage for future enhancements, combining AI precision with Db2's robust architecture to deliver unprecedented query optimization capabilities for modern data workloads.
