View Printable Version

D14 - The Db2 Parallel Universe

Session Number: 5035
Track: Db2 for LUW - II
Session Type: Podium Presentation
Primary Presenter: John Hornibrook [IBM Canada]
Time: May 02, 2018 (04:55 PM - 05:55 PM)
Room: Commonwealth D

Session Code: D14
Speaker Bio: John is a Senior Technical Staff Member responsible for relational database query optimization on IBM's distributed platforms. This technology is part of Db2 for Linux, UNIX and Windows, Db2 Warehouse, Db2 on Cloud and the Big SQL feature of Big Insights. John also works closely with customers to help them maximize their benefits from IBM's relational DB technology products.
Audience experience level: Intermediate
Presentation Category: Data Warehousing and Business Intelligence, Managing Performance
Presentation Platform: DB2 for Linux, UNIX, Windows
Audiences this presentation will apply to: Data Architects, Database Administrators, New Users
Technical areas this presentation will apply to: Database Performance (DB2 for LUW), SQL and XML Performance
Objective 1: Understand how parallelism works for row-organized processing.
Objective 2: Understand how parallelism works for column-organized processing.
Objective 3: Learn how parallelism works with the database partitioning feature (DPF).
Objective 4: Understand how the optimizer choose parallelization strategies.
Objective 5: Learn what external controls are available to configure query parallelism and how to monitor it.

Abstract:  Db2 is a relational database the can be used for many types of applications, from operational systems running very high number of users and transaction to data marts and large data warehouses. Db2 is particularly good at running complex query and analytical processing on a variety of hardware configurations. Since all modern hardware contains multiple cores and CPUs, and many systems are composed of multiple machines, it is critical that Db2 run individual queries across all available computing resources to meet the ever increasing performance demands of business users. This session will provide details on how Db2 parallelizes query execution in various system configurations, from single node machines to clustered configurations. It will contrast how query parallelism behaves for row and column-organized tables and how to identify parallel operators in the query access plan using the explain facility. Details on how the query optimizer chooses parallelization strategies and the various external controls and monitoring features will also be provided.

For questions or concerns about your event registration, please contact