27

Table and Index Design for Performance

Date: 
Wednesday, 29 lipanj, 2005 - 11:45 - 13:15
Speaker: 
Susan Lawson

This presentation will discuss many issues surrounding alternative table designs and performance. We will look at many different alternatives to help design tables for the best performance possible for the type of application process. There are many issues to consider for high volume insert and update transactions, and a balancing act to be achieved for responsiveness to readers. Table issues such as denormalization, non-column attributes, indicator columns, repeating columns, derived data, aggregated columns, table flipping, identify columns, surrogate keys, horizontal and vertical partitioning. We will also look at issues with designing VLTBs (Very Large Tables) and XVLTBs (eXtremely Large Tables) – the rules have changed! We also look at index challenges.

Creating Tables to Support High Volume Inserts
Alternate Column Designs for Application Performance
Building VTLBs and XVLTBs
Index Impacts and Usage – NPSIs and DPSIs

I.Alternative table designs and performance.
a.Design tables for the best performance possible for the type of application process.
b.Issues to consider for high volume insert and update transactions
c.Performing a balancing act to be achieved for responsiveness to readers.

II.Table issues
a.Denormalization
b.Non-column attributes
c.Indicator columns
d.Repeating columns
e.Derived data
f.Aggregated columns
g.Table flipping
h.Identity columns
i.Surrogate keys
j.Horizontal and vertical partitioning
k.Declared Temporary Tables

III.Designing VLTBs (Very Large Tables) and XVLTBs (eXtremely Large Tables)
a.Issues that are different from traditional design
b.What works and what does not
c.Purging and Archiving Issues
d.Using Additional Tables for Access
e.Actual client experience stories

IV.Index Issues
a.Challenges with NPSIs
b.Challenges with DPSIs
c.Uniqueness and Clustering

Bufferpool design anno 2005

Date: 
Wednesday, 29 lipanj, 2005 - 11:00 - 11:45
Speaker: 
Klaas Brant

Buffer pools and there settings is the #1 tuning opportunity in DB2. But there are many settings and there so many stories about buffer pools that it sometimes not clear where to begin and what to expect as result. Also when you move to version 8 options like hiper pool and data space no longer exist. Now what? Klaas will tell you how to prepare for version 8 with a practical working solution in version 7. Because the golden rule still is: your DB2 performs better if the buffer pool setting are tuned.
Topics: What are buffer pools and why are they important, one of more buffer pools?, getting rid of the hiper pool, monitoring the efficiency

DB2 Triggers - An Overview

Date: 
Wednesday, 29 lipanj, 2005 - 10:00 - 10:45
Speaker: 
Phil Grainger

Triggers have now been around for some time in DB2 for z/OS so perhaps it is time for a brief refresher. In this session, Phil will review the trigger mechanism of DB2 for z/OS and will make some recommendations on trigger usage. As well as simple trigger creation, the uses of transition variables and transition data will be covered. In addition, the performance enhancements of DB2 Version 8 will also be reviewed.

Query Optimization Part 2

Date: 
Wednesday, 29 lipanj, 2005 - 09:00 - 10:00
Speaker: 
Namik Hrle

This presentation discusses the fundamentals of the DB2 query optimization techniques. Part 1 gives an overview of the basic SQL constructs and explains their impact on query performance. The DB2 runtime architecture and impact of predicate evaluation order and timing are described. The presentation also gives an introduction to the basic optimization algorithm. Part 2 discusses query transformation, including techniques that DB2 employs and how these transformation techniques impact query performance. An overview of cost estimation of access paths is included. Finally, the presentation explains how data statistics can impact the optimal access path that DB2 selects.

DB2 Database Maintenance on Demand – Enabled with RTS

Date: 
Tuesday, 28 lipanj, 2005 - 17:00 - 17:45
Speaker: 
Ulf Heinrich

Real 24x7 availability requires real-time database maintenance that guarantees security, reliability, and top performance.  This presentation will show you how to exploit the online capabilities of IBM’s Utility Suite in conjunction with real-time statistics to shift database maintenance from your shrinking batch window into a time period during which the utility execution is really needed.

Bullets:
1.Background: Current state of database maintenance strategies
2.DB2 catalog statistics vs. real-time statistics - what’s the difference?
3.Real-time 24x7 database maintenance - a detailed look
4.Real-time extent management - how it eliminates system outages
5.Real world experiences - real-time maintenance results

Stored Procedures

Date: 
Tuesday, 28 lipanj, 2005 - 16:00 - 17:00
Speaker: 
Philippe Herman

Although Stored Procedures are starting to become established in some application architectures, it is still considered by many as something either to new or not valuable.
This session will provides a walk through into the world of the DB2 stored procedures.
We will focus on

The environment setup needed to run Stored Procedures
The performance aspects. (When to use them and when to avoid them)
The 'political' issues from a DBA point of view?
Some interesting IBM supplied stored procedures.
 

Top 20 Performance Myths 2005

Date: 
Tuesday, 28 lipanj, 2005 - 14:15 - 15:45
Speaker: 
Susan Lawson

Abstract:
DB2 appeared in 1983, and standards and guidelines were soon to follow. This led to a lot of rules and standards in the beginning, and some have become so ingrained that they became almost by-laws of systems development.  We need to take a hard look at these guidelines and how they may not apply in today’s high performance environments and may even be hurting performance.
We are going to look information passed on through the years and evaluate its applicability today. There are no two systems exactly the same where everything will be done exactly the same.  Sure, some myths may appear to be nice starting points and perhaps some of them are. The reason is because this particular method may have worked and produced adequate results in this particular environment, but surely is not appropriate for all environments and could even prove harmful. It all goes back to the saying “Don’t believe everything you hear (or read)’. Take all implementation and performance recommendations with an open, yet cautious mind, and then evaluate them against your environment. This presentation quick look at what we have seen as the top 20 performance myths and will include new material up to date with Version 8.

Bullet Points:
1.What are DB2 Performance Myths?
2.Where did they originate?
3.How do I identify if they are causing me problems?
4.How do I fix the problems?
5.How do I avoid falling victim to future performance problems?.

Query Optimization Part 1

Date: 
Tuesday, 28 lipanj, 2005 - 12:15 - 13:15
Speaker: 
Namik Hrle

This presentation discusses the fundamentals of the DB2 query optimization techniques. Part 1 gives an overview of the basic SQL constructs and explains their impact on query performance. The DB2 runtime architecture and impact of predicate evaluation order and timing are described. The presentation also gives an introduction to the basic optimization algorithm. Part 2 discusses query transformation, including techniques that DB2 employs and how these transformation techniques impact query performance. An overview of cost estimation of access paths is included. Finally, the presentation explains how data statistics can impact the optimal access path that DB2 selects.

Java and the Wild Wild Web - Crash Course No.2

Date: 
Tuesday, 28 lipanj, 2005 - 10:45 - 12:15
Speaker: 
Maria Sarikos

This is the sequel of the award winning presentation - Java & the Wild Wild Web. In this presentation we delve into more advanced Web topics such as – The ABC’s of creating a Java pgm accessing DB2 ; Web services and their components and how they interact with DB2; Why XML is so important and how DB2 handles it; Why SQLJ is so much better for you & faster than JDBC; Plus we review some Web Performance commandments.

Presentation Bullet Point:
1. The ABC’s of creating a Java pgm accessing DB2
2. Web services and their components
3. XML - why its so important & how DB2 deals with it
4. Why SQLJ is better for your and your DB2’s health
5. Web Performance commandments

Online Reorg, almost perfect

Date: 
Tuesday, 28 lipanj, 2005 - 09:45 - 10:35
Speaker: 
Klaas Brant

Online reorg has been with us for several releases now. But with every release IBM has been improving this utility. So things were and are not always perfect. Let’s review again how online reorg works and what has been changed since the introduction. Finally Klaas will review which issues still exist and will be changed in the next release.
Topics: How online reorg works, your perfect reorg run, issues fixed and issues to be fixed

Syndicate content  

sqladria

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.

Follow us

facebook
linkid
twitter

 

Events in 2023

VODICE -  12th - 15th June 2023