Tuning DB2 SQL for Performance
Sheryl  M. Larsen, Sheryl M. Larsen, Inc., SMLI

This class will begin with a review of the SQL coding performance rules. The first half of the class will concentrate on the skills required to be competent at changing the DB2 optimizers mind: what each access path is, when they are good, when they need to be altered, reading PLAN_TABLE , Visual Explain, and applying the recommended tuning techniques.  The second half will focus on how to determine if a query has a non-optimal access path.  A couple of estimation techniques will be demonstrated.  For the final test, a poorly coded query will be provided for the students to fix. 

Sheryl M. Larsen, Inc., SMLI, is a firm specializing in Advanced DB2 Consulting and Education (www.smlsql.com). Sheryl M. Larsen is the founder/president and is an internationally recognized researcher, consultant and lecturer and is known for her extensive expertise in SQL. She has over 20 years experience in DB2, published many articles, several popular DB2 posters, and co-authored a book, DB2 Answers, Osborne-McGraw-Hill, 1999, and will co-author a new book, in 2006 Design Patterns: Data Strategies for Performance. She is frequently asked to keynote DB2 world wide events. She is an IDUG "Hall of Fame" speaker, President of the Midwest Database Users Group (www.mwdug.org), and a member of IBM's DB2 Gold Consultants program. She can be reached at SherylMLarsen@cs.com or (630) 399-3330.

 

Performance for SOA Java DB2 Applications
David Beulke

David Beulke is an internationally recognized DB2 consultant, author and lecturer. He is known for his extensive expertise in database performance, data warehouses and internet applications. He is currently a member of the IBM DB2 Gold Consultant program, co-author of the IBM V8 and V7 z/OS DB2 Certification exams, co-author of the Business Intelligence Certification exam, past president of the International DB2 Users Group (IDUG), columnist for DB2 Magazine, former instructor for The Data Warehouse Institute (TDWI), and former editor of the IDUG Solutions Journal. He has helped clients with system and application issues on the mainframe, UNIX and Windows environments and has over 20 years experience in numerous industries. He can be reached at DaveBeulke@cs.com or 703 798-3283.

In the Performance for SOA Java DB2 Applications class the application DBA and developer learn the common SOA Java performance pitfalls and how to avoid them.  This class takes the attendee through the application architecture and discusses the Java application pattern choices.  The class highlights the common pitfalls and how to avoid performance problems related to Java sessions, messages and application persistence issues.  The class also highlights the best practices for database connections, SQL access and error checking within a high performance environment.  Discussing these areas along with common Java JSP and servlet programming traps and debugging can help your application developers improve your application scalability and overall performance

Outline:

 

SOA Java Architecture

 

·          Java Frameworks

 

·          Application Patterns

 

·          Common Components

 

·          Technology Components

 

Servlets and JSPs, Considerations

 

·          Model View Controller Pattern

 

·          Servlet Programming Patterns

 

·          Servlet Performance Considerations

 

·          JSP Programming Patterns

 

·          JSP Performance Considerations

 

EJBs Considerations

 

·          EJB with Model View Controller Pattern

 

·          Programming efficient EJB

 

·          Session and persistence with EJBs

 

·          Memory Leaks

 

SOA and Java Scalability

 

·          Patterns for Performance

 

·          Reuse and object oriented programming

 

·          Web Services considerations

 

·          Trends, Fades and Reality

 

Java Application Database Considerations

 

·          Model View Controller Pattern with EJB

 

·          Programming efficient EJB

 

·          Session and persistence with EJBs

 

Finding the performance problems

 

·          Server Considerations

 

·          Logging

 

·          Profiling

 

·          Application Tracing