Performance problems can be tough to solve, especially if you've not had to deal with them much before. In this session, we'll go through a step-by-step, decision tree-based approach for tackling performance problems. We start by identifying the core area of the bottleneck - is it time spent processing, or time spent just waiting? Then, by using DB2's powerful monitoring interfaces, we refine our diagnosis, drilling down layer by layer. This logical, efficient approach not only helps us get to root cause, but it does so in a way that avoids wasting time by "trying random stuff" which isn't likely to help. Based on a prior & highly successul IDUG session of the same name, this presentation has been completely revised and updated, using the latest levels of DB2 LUW.
Steve Rees is an IBM Senior Technical Staff Member focusing on DB2 LUW performance. Steve joined the IBM Toronto Laboratory in 1991, and has worked on DB2 LUW ever since. He currently leads teams across multiple areas of DB2 performance, including pureScale and SQLDB transactional cloud, as well as Optim tools performance. Steve has an ongoing passion for developing best practices in the area of monitoring and tuning DB2, and presents regularly at conferences.