DB2 for zOS V8
Platform: OS/390 and z/OS - Audience: All

This presentation will provide an overview of features/enhancements to be found in the recently announced DB2 for z/OS V8.

Hugh Smith is the leader of the distributed and client/server measurement team of the DB2 UDB for OS/390 and zOS Performance department. He specializes in connectivity and performance tuning issues of such environments. He also provides consulting services for customers in these areas. He has been in the job since 1992. Prior to that, he held positions in application generator development, CICS support, and database product planning.


TO SCORE OR NOT TO SCORE: THAT IS THE QUESTION
Platform: Linux, Unix, Windows - Audience: Intermediate to Advanced

Customer Value Management (CVM) is a vision and CVM Analytical Environment is the datamart to execute this vision. The development of the new 5 TB datamart gave us the opportunity to test an assortment of DB2 UDB EEE features, use many of them and not use a few. This rich BI environment uses a variety of tools and techniques including IBM Intelligent Miner Scoring Services, AbIntio, Microstrategy products, DB2 OLAP functions and horizontal partitioning of very large tables.

Sonina has 17 years experience in data modeling and database administration. She is currently leading a group of 10 intermediate and senior DBAs, that provide support for 9 Data Warehouse/Datamart environments with 100GB to TB of data for Bank of Montreal.

TRIGGERS AND OTHER V7 TALES
Platform: OS/390 and z/OS - Audience: Intermediate to Advanced

Hari's back from his US adventures with his personal perspectives on implementation, usage and testing of some DB2 for OS/390 and z/OS V7 elements:
-Multi-edit triggers - getting more bang out of your triggers
-Trigger performance considerations and pitfalls
-To trigger or not to trigger... alternatives
-A trigger by any other name... - naming conventions
-Comedy of Errors - trigger testing and what happens with various types of error conditions
-Triggering Stored Procedures - some considerations
-Three amigos - DSNTIAUL, V7 UNLOAD, High Performance Unload – some considerations, including the mystery of the duplicate unload records
-A key! A key! A kingdom for a key!
-Various logical, historical, recovery, debug and philosophical issues/trade-offs

Hari is an international expert in DB2 for OS/390 performance, design and exploitation with consulting engagements in Canada and the US. He is a past speaker at the DB2 Technical Conference, IDUG, the IDUG Technical Symposium, the TRIDEX DB2 User Group in New York City and a regular speaker at the Central Canada DB2 User Group. He has been an IBM panelist at the DB2 Technical Conference and the IDUG Technical Symposium. Prior to incorporating, Hari was for many years IBM Canada Ltd.'s country level DB2 for OS/390 expert.

NEW DATA CLUSTERING TECHNIQUES FOR PERFORMANCE: MULTIDIMENSIONAL CLUSTERING
Platform: Linux, Unix, Windows - Audience: All

This presentation will present in detail this very new and exciting feature, which is unique to DB2 for Unix and Windows platforms(so far), and to the industry in general. It is a high availability and high performance enhancement for large to very large databases, especially useful in data warehouse environments.

For the past 5 years, Lesley has been working as a Senior developer on the Index Management team, as has designed and implemented clustering and bidirectional indexes, along with other performance features. Most recently, she has been working on designing and developing the latest data clustering technology for DB2 UDB, multidimensional clustering Lesley has implemented a significant portion of this feature, and led the development team for this project.