Skip to main content

BI Best Practices Technical Chat Program

Information Management Training

These complimentary Chat with the Lab programs are brought to you by a combination of the Business Intelligence Best Practices Team and The Toronto and Santa Teresa Labs. Download one or all of the chats at no charge! Chat is a technical program designed for DBAs, system administrators and architects interested in the latest and greatest technical information regarding DB2 DPF specifically within a Business Intelligence environment. The program was designed to provide our customers a technical forum to:

Chats run 1-2 times per month via audio conference and webinar. If you are interested in being a member of the "Chat List" where you will receive invitations for future chats, please send an email request today!

Order a CD

NEW BI Best Practices Chat CD (vol 1): 2003-2004 BI Chats
This CD is available at a minimal cost, less than 1 USD. Part number: GK4T-0467-00 To order, click here.

Chat with the Lab - Workload Management with Query Patroller
04/20/06
Speaker - Dwaine Snow and Paul Zikopoulos
DB2 Query Patroller is a powerful query management system that you can use to proactively and dynamically control the flow of queries against your DB2 data warehouse database. The features of Query Patroller allow you to regulate your database's query workload so that small queries and high-priority queries can run promptly and your system resources are used efficiently. In addition, information about completed queries can be collected and analyzed to determine trends across queries, heavy users, and frequently used tables and indexes. This session will take a closer look at the features, architecture and reporting capabilities of Query Patroller.


Chat with the Lab - What's new for Information Warehousing with DB2 Viper
03/14/06
Speaker - George Baklarz
Our first chat of 2006 provides a preview of DB2 Viper technology and what DB2 Viper will mean to the Information Warehousing/Business Intelligence Community.


Chat with the Lab - Trends and Developments at the Frontiers of Data Warehouse Scalability
05/19/05
Speaker - Richard Winter and Dr. Bill O'Connell
Richard Winter from Winter Corp. presents his view on trends in the data warehouse market and also a discussion on a very large benchmark which was recently conducted .
by Winter Corp. with DB2 DPF and IBM pSeries hardware. Additionally, Dr. Bill O’Connell, DB2 Chief Architect, Toronto Lab talks about lessons learned from this benchmark, and where DB2 is going to address extreme data warehousing.


Chat with the Lab - Increasing Your Data Warehouse Value Factor Through Consolidation
4/28/2005
Speaker - Dr. Bill O'Connell and Paulo Pereira
Our first Chat with the Lab session of 2005 presents IBM's point of view on building data warehouses for increasing the value factor of information. The main business reasons driving the cost benefits of an integrated and consolidated enterprise data warehouse are discussed. Finally, the technical challenges behind consolidation and how IBM solves such issues with DB2 based solutions are touched upon.


Chat with the Lab - What's New for BI in DB2 UDB V8.2
8/17/2004
Speaker - Glen Sheffield
DB2 UDB V8.3 offers features that significantly increase productivity of database administrators and developers and meet complex data management requirements. The new DB2 Design Advisor, enhancements to the query optimizer and Materialized Query Tables, and analytic capabilities provide more flexibility to solve complex business intelligence challenges, while reducing database administration tasks at the same time. This session presents an overview of the many new and improved features packed into this release.


Chat with the Lab - Federation in BI
7/20/2004
Speaker - Dr. Bill O'Connell
Consolidation versus federation augmentation is always a contentious topic. In this session, consolidation will be briefly discussed, along with federation in warehousing. The focus of this talk will be on federation augmentation, but will touch on architecting guidelines for consolidation. The presentation will overview federation augmentation scenarios in a warehouse, and when federation make sense; example scenarios being deployed will be discussed. To provide a level of appreciation of federation characteristics, performance trade-offs of federated queries, as well as how to read federated query access plans will be discussed.


Chat with the Lab - SMS/DMS Tablespaces
6/17/2004
Speaker - Matt Huras
This talk will provide an overview of the internals of tablespaces in Distributed DB2, guidance on how to setup and configure them, as well as information on upcoming advancements in this area.


Chat with the Lab - DB2 Data Warehouse Backup & Recovery
6/8/2004
Speaker - Dale McInnis
This session covers our current technology for DB2 data warehouse backup and recovery as well as information on where we are going with backup and recovery in the future.


Chat with the Lab - Performance Gains w/DB2 V8
5/20/2004
Speaker - Berni Schiefer
This chat covers the performance enhancements in DB2 UDB Version 8.1, and contrasts the performance of DB2 Version 7 with DB2 version 8 for data warehousing.


Chat with the Lab - DB2 Data Warehouse Edition V8.2
5/11/2004
Speaker - Anant Jhingran, Sriram
DB2 Data Warehouse Edition V8.2 extends the powerful function of DB2 Universal Database with advanced BI features for building and working with data warehouses and data marts. This chat describes DB2 Data Warehouse Edition (DWE) and its value for addressing Business Intelligence needs of customers. In particular, we will preview the new features of DWE in the "Stinger" release timeframe. We will describe the enhancements to the DB2 UDB engine in the Business Intelligence area, new features in DB2 Cube Views, DB2 Intelligent Miner functions, DB2 Query Patroller, and DB2 Warehouse Manager. If time permits, we will demonstrate the capabilities of Cube Views and Intelligent Miner.


Chat with the Lab - Partitioning Options
4/27/2004
Speaker - Dwaine Snow
This chat session presented a discussion of the issues and options in data warehouse partitioning now, optimizing for DPF, & future implications.


Chat with the Lab - Query Patroller
4/6/2004
Speaker - Dwaine Snow
DB2 Query Patroller is a powerful query management system that you can use to proactively and dynamically control the flow of queries against your DB2 database. The features of Query Patroller allow you to regulate your database's query workload so that small queries and high-priority queries can run promptly and your system resources are used efficiently. In addition, information about completed queries can be collected and analyzed to determine trends across queries, heavy users, and frequently used tables and indexes. This session introduces the new Query Patroller features and architecture. It will explain the use of new advanced reports that let you manage the system performance and disk usage.


Chat with the Lab - Data Re-distribution Options
3/23/2004
Speaker - Glen Sheffield
Data partitioning is fundamental to the scalable architecture of DB2 UDB for distributed platforms. But what happens when you add capacity to your system? Do you need to add new partitions and re-distribute your data ? If so - How? This chat session reviews the principles of DB2 data partitioning, planning for system growth, and different re-distribution options (including the DB2 re-distribute utility). We will also discuss reducing the number of partitions for cases where hardware upgrades reduce the number of CPUs needed to handle the data volume.


Chat with the Lab - Understanding and Leveraging Materialized Query Tables (MQTs)
12/11/2003
Speaker - Bill O'Connell
Materialized Query Tables (MQTs), formerly known as Automatic Summary Tables (ASTs) provide a powerful look aside capability for the DB2 optimizer to significantly improve the performance of complex and long running decision support queries over voluminous data. In some cases, the improvement has been in orders of magnitude! MQTs are defined by the DBA (as indices too) and require no SQL re-writing to exploit them. The system automatically will cost routing the query to them automatically. The system also incrementally maintains these MQTs as indices are (both immediate and deferred definitions of them). This session provides an overview on MQTs in DB2 for LUW (Linux/Unix/Windows), includes discussions on matching capabilities, incremental maintenance for insert/update/deletes as well as bulk loading, and finally integration with DB2's OLAP capabilities. This includes both the OLAP and Statistical functions, as well as roll-up and cubing operators.


Chat with the Lab - DB2 Performance - Tips and Best Practices for a Data Warehouse Environment
11/19/2003
Speaker - Haider Rizvi
The industry-leading shared-nothing architecture of the DB2 product allows customers to start small and easily scale-out to bigger databases as needed.This session discusses the performance and scalability characteristics of the DB2 universal database as it pertains to the data warehousing implementations. We also discuss best practices for best I/O performance, how to tune various db2 parameters and settings, as well as some of the new DB2 features, such as MDC tables and MQTs.


Chat with the Lab -DB2 Cube Views
11/04/2003
Speaker - Gary Robinson
IBM provides relational OLAP power in a new add-on package for IBM DB2 Universal Database™, a set of features for multidimensional data management called IBM DB2 Cube Views. DB2 Cube Views is built from the ground up to offer heightened awareness of OLAP in the underlying database catalogs, optimization for OLAP-style data access and new interfaces to support Web services.

DB2 Cube Views enables faster deployment, easier management, cleaner source data and better performance across all analytical applications. It offers rapid, smooth integration with a broad range of OLAP add-ons and specialized analytic tools from IBM Business Partners such as Ascential, Business Objects, Cognos, Metadata Integration Technologies, Microstrategy, Brio, and other leading vendors. This session provides a technical overview of DB2 Cube Views and how you can leverage its integration with partner tools and applications to provide better OLAP solutions faster.


Chat with the Lab - Data Warehouses and Information Integration - Best Practices and Performance Characterizations
10/14/2003
Speaker - Dr. Bill O'Connell
The enterprise data warehouse has a fundamental value proposition that is not altered or diminished. However, warehouses that have a huge investment base are being continually challenged to deliver even more business value. This session will first quickly overview warehousing trends, such as real-time warehousing, consolidation, and federation augmentation. Once briefly overviewing these trends, the augmentation value of federation will be discussed in more detail. Federated technology is being used to extend the enterprise data warehouses to more add customer value…..not to replace the data warehouse. The session will then describe various customer exploitations of federation today, as well as the limits of federation. Finally, the session will provide a characterization of federated performance.


Chat with the Lab - Extending the Data Warehouse through Federation
10/9/2003
Speaker - Dr. Barry Devlin
Today, business needs such as Customer Insight, Real-time Warehousing, and Business Activity Monitoring often demand more than can be delivered with a traditional data warehouse. They require easy, direct or near real-time access to a wide variety of information, much of which is outside today's warehouse. So - is a new approach to data warehousing needed or can the current approach be extended? This session provides an architectural overview of how to extend the traditional data warehouse with direct access to some key sources of vital data. We show how the new concept of Information Integration, or Federation, is a natural extension to data warehousing, provide an introduction to the new DB2 Information Integrator technology and describe some examples of how it can be used to enable a dynamic and federated view over the entire information of the organization, where the data warehouse remains a key component.


Chat with the Lab - Multi-Dimensional Clustering
9/29/2003
Speaker - Leslie Cranston
(MDC) is a unique, patented feature in DB2® Universal Database Version 8.1. MDC enables a table to be physically clustered on more than one key, or dimension, simultaneously. Properly implementing MDC tables can result in significant improvements in the performance of queries, as well as significant reduction in the overhead of data maintenance operations such as reorganization, and index maintenance operations during insert, update and delete operations. In order to take full advantage of the performance benefits that Multidimensional Clustering of data has to offer, it is important to fully understand the factors involved in selecting appropriate dimensions BEFORE you create your multidimensional clustering tables. This session describes the considerations that should be made when choosing appropriate dimensions, block size, and generated columns in order to create optimal multidimensional clustering tables. We will also look at some MDC performance results and performance tuning tips, and finally look at considerations involved in choosing between different the data clustering and partitioning techniques available.