SQL can be processed in parallel by DB2. How does this work? What is the difference between IO parallelism and CPU parallelism? How do I ask for parallelism and how can I find out it has will be or has been used? Which settings and parameters are important? Klaas will show you that processing SQL can be a great feature of DB2 but there are cases when parallelism is very bad option. Topics: Types of parallelism, asking for parallelism (degree), parameters that influence parallelism (Bufferpools, ZPARM, SQL), Monitoring and predicting (Explain and accounting trace), Rules for using parallelism
This presentation shows the real and often surprising results of analysing DB2 environments in representative DB2 for OS/390 installations (Banks, Insurance, Industry and Cross Service Companies) in Europe and USA.The most common discrepancies and inconsistencies affecting the availability, the performance and the reliability of DB2 will be reported as well as their reasons, impacts and possible ways to solve or palliate them.
PRESENTATION OUTLINE:
Part 1: BACKGROUND: Presentation about the investigations done in order to detect discrepancies and inconsistencies affecting the following areas of concern: Catalog size Catalog reliability and consistency Space saving Application/utility Concurrency CPU and IO performance Resource availability and data recoverability Parallel-sysplex datasharing DB2 Upgrades Security handling Change management and housekeeping
Part 2: FINDINGS: Presentation of the discrepancies and inconsistencies encountered in representative DB2 for OS/390 installations (Banks, Insurance, Industry and Cross Service Companies) in Europe and USA.
CONCLUSION/SUMMARY: Report of 'very often', 'often' and 'rarely' encountered discrepancies and inconsistencies and a rating of the benefits of resolving them.
Abstract: This presentation will cover aspects of Version 8 of DB2 on the z/OS platform and how they will impact our applications, SQL and performance. There are several new constructs in DB2 in the next release such as materialized query tables, multi-row fetch/insert, dynamic scrollable cursors, imbedded inserts, better matching predicates, sequence objects and more. There have also been many improvements DB2s usage of indexes and this will have an impact on how we code our SQL. WLM is playing more of a part in our environments and we need to begin the migration toward it if we have not. We need to review these features and see how they impact current applications and help us prepare for the future in order to get the best performance possible from our DB2 applications
Five Bullet Points Overview of V8 Features that SQL and Applications Discussion of New SQL Functionality Discussion of New Application Programming Features Discussion of Why Its is Important to Use The New Features Discussion of How to Plan and Best Implement New Features
Outline: Application and SQL Materialized Query Tables Coding Predicates for DPSIs Table Function Cardinality Multi-row FETCH and INSERT DDF Block Fetch INSERT within SELECT Dynamic Scrollable Cursors Identity Column Enhancements Sequence Objects Indexable Unlike Types Increased of Number of Tables in a Statement Stored Procedures and WLM
Today’s application environments demand more availability and high performance that we ever had in the past. One problem with achieving both of these is locking. Locking can hold resources that effects availability and can slow down performance. We will take a look at some things we can do to minimize this without compromising data integrity. We will also look at application availability issues, because if you are not available….you are not performing!
Locks and Lock Avoidance Bind Parameters Uncommitted Read Release Deallocate Current Data Lock Avoidance How it works Application Design Designing for no deadlocks Restructuring code for maximum throughput Commit Strategies Index availability Cursor Usage
DB2 for z/OS plays an integral role in any enterprise network computing solution. This presentation discusses what is new, hot, and on the horizon in the overall area of network computing in DB2. Topics covered include connectivity from ODBC, Java, and .net clients, what is new in Version 8 to support these applications, plus information on web services access and XML support for DB2 for z/OS applications.
Part1 SQL Performance is key to the success of any application, however programmers are often not given the full story and hence queries are not coded in the most optimal way. This session is here to remedy this situation. Part 1 deals with gaining solid in-depth understanding of DB2 EXPLAIN. Filter Factors and Runstats are closely examined and insight is gained into which statistics columns trigger certain optimizer decisions.
Bullet Points Why is SQL Performance so Important? DB2 Explain – The Hieroglyphics translated Understanding Filter Factors Runstats & the Statistics Tables Summary
Part2 In this presentation we take a good look at the optimizer and how it goes about creating access paths. The different types of Joins are examined and the reasons why DB2 chooses certain joins over others. Inner and all the flavours of Outer joins are discussed and how they compare performance-wise to other SQL techniques. And most importantly SQL Coding Do's and Don'ts are examined.
Bullet Points The Optimizer – Understanding how it thinks! Joins – The different types, how they work More on Joins - Inner, Outer, Left & Right Outer Speeding up Performance – The Do’s and Don’ts Conclusion
One of the major changes to DB2 for z/OS Version 8 is in the way partitioned tablespaces are created and managed. This presentation will introduce you to table-based partitioning, all of the new features that are available and some pointers on how these enhancements will benefit application design and operation. Unfortunately, there are one or two scares along the way, but these are easily understood and by the end of the session, Version 8 partitioning should hold no fears!
Release 11 of the Unicenter Database Management tools for DB2 for z/OS is the first release of tools from CA (indeed from any major software vendor) that supports DB2 UDB for z/OS Version 8. However, there is much more to r11 than Version 8 support. In this session, Phil will summarise many of the key enhancements to our DB2 tools.
European SQLAdria Seminar – Vodice 2024
SQL Adria is the independent, non-profit organization that gathers relational database users for Croatia and Slovenia. It was founded 1994. and in the same year it has become the regional user group.
8 reasons to come Speakers Events Contact
VODICE - 10th - 13th June 2024