View Printable Version

E10 - Benefits of Apache Spark on Z Systems

Session Number: 3306
Track: App Dev
Session Type: Podium Presentation
Primary Presenter: George Wang [IBM ]

San Antonio => Wed, May 25, 2016 (09:15 AM - 10:15 AM)

Speaker Bio: George Wang is a software engineer at IBM DB2 for z/OS Development from Silicon Valley Lab in California. He is the technical lab advocate and Spark liaison for large banking customers on z system. His technical expertise focuses in architecture design of core system engine components for developing high availability features and providing solutions for warehouse applications in support of high volume online transaction database processing. He also focuses to enable the integration of analytics with transaction environments with the adoption of Apache Spark for z systems for advanced analytics capability on z data.
Audience experience level: Beginner, Intermediate, Advanced
Presentation Category: Application Design, Data Management, Big Data
Presentation Platform: Cross Platform
Audiences this presentation will apply to: Application Developers, Data Architects, Database Administrators, Systems Programmers, New Users, IT Managers
Technical areas this presentation will apply to: Database Performance (DB2 for z/OS), High Availability (DB2 for z/OS), New Release (DB2 for z/OS), SQL and XML Features, SQL and XML Performance, User Experiences
Objective 1: z system platform is a unique analytic implementation. The talk will cover what are the analytic opportunities and challenges
Objective 2: What's the IBM strategy and added value to Apache Spark, and why spark matters to the z ecosystem will be discussed in this session
Objective 3: Using Apache Spark for the IBM Analytics Platform is to be showcased with a use-case demo

Abstract:  The recent growth and adoption of Apache Spark as an analytics framework and platform is timely and helps meet these challenging demands. This presentation showcases that Apache Spark can be used in conjunction with high volume OLTP database technologies such as IMS, DB2 for z/OS, to perform analytical tasks on vast amounts of data without impacting transactional performance. This presentation also provides a scenario on how to provide this capability for users running XML analytics using the Spark framework with SparkSQL. The concept makes the use of the in-memory clustered computing without storing the data outside of an existing database.

For questions or concerns about your event registration, please contact