IDUG NA 2026

This session presents a comprehensive approach to database backup and recovery using Snapshot Backup technology integrated with Cloud infrastructure. The solution leverages the Db2 ACS Scripted interface to automate backup operations on Linux systems, enabling efficient data protection strategies for enterprise databases. The implementation utilizes Linux LVM (Logical Volume Manager) to create consistent point-in-time snapshots of database volumes, ensuring data integrity during backup operations. By combining snapshot-based backup techniques with cloud object storage capabilities, this approach delivers a robust disaster recovery solution that minimizes backup windows, reduces storage costs, and enhances data availability. The scripted automation through Db2 ACS interface ensures consistent, repeatable backup procedures while maintaining operational efficiency in Linux environments. This methodology addresses critical requirements for modern database management, including rapid recovery time objectives (RTO), minimal recovery point objectives (RPO), and compliance with data retention policies through cloud-based archival strategies.

The folks at the Db2 Lab provide us with a great tool to monitor Db2. It's free and already installed with Db2. This session will show you how db2mon works and how to use it. Then, we'll discuss how to run it in different ways for different use cases. Next, we'll present a way to run db2mon for historical data gathering and trend analysis. Finally, we'll show examples of the data db2mon provides and what to look for in various performance scenarios.

Not all indexes are created equal. We explore the architectural differences between Partitioned, Non-Partitioned (NPSI), and Data-Partitioned (DPSI) indexes and when to deploy each for performance. Crucially, we explore each type's optimal recovery strategy so you can stop guessing between Rebuild and Recover.

A review a of recent enhancements to the BMC Db2 High Speed Utilities

Fast copies of Db2 for z/OS tables are increasingly difficult as newer features add structural complexity. Table versioning, partition growth and rotation, XML and LOB data, temporal and archive relationships, and deferred VSAM creation can prevent direct dataset-level copies and force slower fallback methods. This webinar explains the key challenges and what they mean for copy speed and reliability.

Efficient data extraction is critical for analytics, migrations, and modern application integration. Optim High Performance Unload (HPU) provides a powerful, flexible, and lightning-fast way to unload Db2 data without impacting production workloads. This session will explore how HPU can be used for faster unloading and migrating data by using direct access methods to deliver unmatched performance compared to traditional unload utilities. In addition, we will demonstrate how HPU can recover data directly from backup images in the event of table loss, ensuring business continuity and minimizing downtime. Attendees will learn best practices for optimizing unload tasks, and handling large-scale migrations to cloud or hybrid environments. Real-world case studies will highlight performance gains, reduced disk space consumption, and simplified operations. Whether you are a DBA, or architect, this session will equip you with practical strategies to maximize the value of HPU in your enterprise.

This 4 hours workshop - knowledge transfer session will be conducted by following presenters: Toine Michielse, Michal Bialecki, Emil Kotrc, Denis Tronin
In the world of Db2 for z/OS, "set it and forget it" is a recipe for performance degradation and wasted storage. As data volumes explode and batch windows shrink, traditional housekeeping—REORG, RUNSTATS, and IMAGE COPY—must evolve from manual chores into a streamlined, automated strategy.
This deep-dive workshop provides a comprehensive blueprint for modern Db2 maintenance. We will move past the basics to explore how to balance object availability with optimal performance. Attendees will learn how to leverage the latest utility enhancements to reduce CPU consumption, manage multi-terabyte objects, and ensure that the optimizer always has the "ground truth" through intelligent statistics collection.