Day 1

Winning with Machine Learning. Monetize the data behind your firewall

DanielHernandez.jpg

Daniel Hernandez, VP IBM Analytics, IBM

Tuesday,  October 3rd  

Competitive pressures and the need to innovate are everywhere. Data is the new currency and exploiting your data to gain insights becomes core to every business. In this presentation, you will learn about IBM's point of view, how we help you progress your journey towards cognitive self-service analytics with Machine Learning, and generate value today and tomorrow, as your organization embraces hybrid cloud solutions.

Biography:

Daniel is currently the offering management leader for the IBM Analytics division and part of a talented team that reinvents the Analytics by focusing on business - building, operating, and enhancing integrated analytic solutions. Daniel always looks beyond the envelope, loves challenges and grows remarkable teams that put the client first.

Expand DB2 Beyond SQL for Advanced Analytics With Embedded Apache Spark

Torsten Steinbach
Software Architect, IBM
  Bio

Torsten Steinbach

Torsten is a software architect at IBM with many years of DB2 product development experience. He worked on DB2 performance tooling and workload management in the past. Today he leads advanced analytics in IBM's data warehouse offerings.

While DB2 is well equipped to drive descriptive analytics such as BI reporting with its powerful SQL capabilities, it cannot handle predictive and prescriptive analytic workloads, until now. Learn in this session how we augment DB2 by coupling it very tightly with an integrated Apache Spark engine that runs collocated and with your DB2 database partitions and is highly optimised for such deployment. We will demonstrate the capabilities live using dashDB local, a docker-based deployment form factor of DB2 that provides this integration out of the box. We will show you how you can use this as a data scientist interactively as well as the straight forward deployment of analytic jobs, models and pipelines into the database and how they can be called via SQL and REST APIs from your applications.

Machine Learning on z for Beginners

Nigel Slinger
Distinguished Engineer, IBM
  Bio

Nigel Slinger

Nigel Slinger is an IBM Distinguished Engineer working in DB2 for z/OS Development. He currently leads the World Wide Technical support team providing L2 support to customers. He is regarded as one of the ultimate authorities on overall DB2 problem determination especially in the DB2 storage manager area which he helped to design.

Machine learning has been around for a long time but now it’s coming to the z/Series machines. The idea of having the computing power go to the data instead of the data to the CPU is an attractive proposition. It allows for more current data, less data movement and more online use of the data. This presentation will cover the basic concepts of machine learning and also the recent advances from IBM on z/Series to accommodate machine learning to ingest data on this platform.

Machine Learning By Examples

Emil Kotrc
Principal Software Engineer, CA Technologies
  Bio

Emil Kotrc

Emil Kotrc is a principal software engineer at CA Technologies working in the Prague Technology Center since 2005. He started on Resource management products for Mainframe and later moved into Database management for DB2 for z/OS development. Before joining CA Technologies, Emil worked in an academic environment. Emil is an IBM Champion for 2015 and 2016.

Machine Learning (ML) is a rapidly growing field of the computer science that has far reaching applications even outside the academics and specialized industries. Thanks to the Big Data (r)evolution, Machine Learning is becoming popular even in smaller businesses. Moreover, there are many tools and projects that allow running ML algorithms even by enthusiasts without a mathematical or statistical background.

In this presentation we will show some of the most common ML algorithms together with their prevalent use cases and we will demonstrate them on very simple examples. We will also discuss a possible integration with DB2. .

Using LUW with DSX (Data Science Experience) to Solve Practical Machine Learning Problems

Mark Ryan
IBM
  Bio

Mark Ryan

I have been part of the team delivering DB2 LUW for 20 years, with leadership roles in documentation, quality assurance, development, and most recently my role is senior manager for DB2 LUW Support. In this capacity I lead a worldwide team responsible for DB2 Support. My current areas of interest include: - self-help for DB2 LUW customers - applying machine learning to better understand customers' experience with DB2 and to anticipate future challenges and opportunities - DB2 LUW as part of a complete hybrid ecosystem, including DSX, DSM, dashDB

This presentation describes an end-to-end example of using Data Science Experience (DSX) with DB2 LUW to solve a practical machine learning problem (analysis of DB2 PMRs / APAR metadata to explain differences in customer experience between DB2 10.1/10.5 and DB2 V11). I will start with an overview of the problem I am trying to solve and then give a quick overview of DSX. The meat of the presentation will be a description (including demo) of the steps taken to exploit DSX with DB2 as the data source to solve the problem. The audience will gain an understanding of the benefits of exploiting DSX with DB2, as well as specific, practical steps to follow in order to realise these benefits.




Day 2

Integration of Apache Spark/Hadoop for z Analytics

Eberhard Hechler
Executive Architect, IBM
  Bio

Eberhard Hechler

Eberhard is an Executive Architect working at the IBM Germany R&D Lab. He is a member of IBM DB2 Analytics Accelerator development. After 2,5 years at the IBM Kingston Development Lab in New York, he worked in software development, DB2 performance, IT/solution architecture, Hadoop/Spark integration and MDM. From 2011 to 2014, he was at IBM Singapore, working as the Lead Big Data Architect in IBM’s Communications Sector. Eberhard is a member of the IBM Academy of Technology Leadership Team, and co-authored the following books: - Enterprise Master Data Management, Pearson plc, 2008, ISBN-10: 0132366258 - The Art of Enterprise Information Architecture, Pearson plc, 2010, ISBN-10: 0137035713 - Beyond Big Data – Using Social MDM to Drive Deep Customer Insight, Pearson plc, 2014, ISBN-10: 013350980X

Exposing z/OS data into Spark to enable data scientist related tasks is becoming increasingly important. This presentation describes integration scenarios between DB2 for z/OS and DB2 Analytics Accelerator with Spark on z/OS and BigInsights for Apache Hadoop and Spark. We demonstrate how to complement traditional z Analytics with Big Data analytics, e.g. by using SQL on open source (Big SQL and Spark SQL), and illustrate the integration of key Spark components into z Analytics with DB2 for z/OS.

IBM Watson Data Platform 101 for the DB2 DBA

James Sobieski
Fourth Millennium Technologies
  Bio

James Sobieski

James Sobieski is a Database expert having worked on every version of DB2 for LUW since DB2/6000 v1 in 1994 and IBM Systems since 1979. He has extensive experience in Data Warehousing, Business Analytics, design and planning, ETL and DB performance and tuning. James is an IBM Gold Consultant and a five year IBM Data Champion. He consults internationally on DB2, Netezza, Big Data, ETL and MDM projects and has consulted with over 300 clients. For 9 years at NASA he optimized the Shuttle Mission Control systems and Payload Data Management system. He led the IBM DB2 Toronto Lab’s DB2 DPF benchmarking team. James scripted a database migration, executing 225,000+ load jobs to move 152 Billion rows in 22 hours. He has 14 IBM certifications and has spoken at over 20 conferences.

IBM’s Watson Data Platform is a powerful technology, architected to provide easy access to dashDB, Watson Analytics, Cognos Analytics, the Data Science Experience, IBM Bluemix Data Connect (ETL), and Predictive and Prescriptive Analytics. This session explains, in DBA terms, the basics to using the major components of the WDP and how to quickly fit the pieces together in a powerful WDP architecture. It covers loading data from on-Prem and public data sources and analyzing that data.

IBM BLU for Spark - an Event store for the next generation of Applications

Namik Hrle
IBM Fellow
  Bio

Namik Hrle
Namik Hrle is IBM Fellow (the highest technical distinction in IBM) and works in the IBM Boeblingen Development Laboratory. He is CTO for IBM Analytics Private Cloud Platform and z Analytics. As IBM Fellow and a member of the IBM Academy of Technology he belongs to a small circle of the top technical leaders whose work and expertise impact the direction of IBM. Holder of numerous patents (he has a title of 'IBM Master Inventor'), outstanding technical achievements, author recognition and corporate awards, Namik Hrle is respected as the ultimate expert in the area of intersection between enterprise applications and information management technologies. He is one of the most sought after experts by customers, IBM sales, marketing and technical support teams world-wide.

This presentation provides a deep dive into the next generation of IBM data store for handling real time event applications from IoT to new Event Sourcing applications. The Store is built on the Open source Spark platform and Object storage and can ingest Millions of transactions per second and provide Highspeed analytics on transactional data in real time. Perfect for Event sourcing applications that need the velocity and volume of data this platform can handle and for Structured Data Lake applications such as the Internet of things.

DB2 for z/OS Modern application: Watson, ML, REST, Spark and more

Jane Man
Senior Software Engineer, IBM
  Bio

Jane Man

Jane Man is a Senior Software Engineer in DB2 for z/OS development. Jane has more than 16 years’ experience in developing high quality data/content-centric solutions for enterprise customers. Jane has authored many patents and whitepapers on current technology topics and is a regular speaker in global conferences. Jane is an IBM Certified System Administrator for WebSphere Application Server, IBM Certified Database Administrator for DB2 Universal Database for z/OS, Linux, UNIX and Windows, IBM Certified Solution Designer for DB2 Content Manager, IBM Certified Deployment Professional for Tivoli Storage Manager, and IBM Certified Application Developer for DB2 Universal Database Family.

This session focuses on implementing modern applications around DB2 for z/OS using the latest technologies. We will discuss how to build a cognitive DB2 solution using Waston, how to do Machine Learning on DB2 data without a GUI, how to do Spark analytics, how to use DB2 as REST provider using DB2 native REST service, and how to use DB2 as REST consumer using JSON and XML features in DB2. Working examples, hints, and tips will be given that you can try out right away!!

Where does DB2 stand in this Big Data World?

Ludovic Jannsens
AE
  Bio

Ludovic Jannsens

Ludovic Janssens has more than 10 years experience on DB2 for z/OS. He is active in the IDUG Content Committee for a few years and has been recognized as IBM Champion for Analytics. Next to DB2 he takes great interest in Data Management architecture and how this could serve the business needs.

DB2 has been for a long time the central data vault for many enterprises. Nowadays, with the advent of Hadoop, NoSQL and other data infrastructures, it's importance appears to decline, but this is not the case. On the contrary, thanks to the new data evolution we can finally clearly define and argue that DB2 can continue to play the beating heart of an Enterprise.


Apache Spark Hands On lab

The Apache Spark Hands On Lab (HOL) will be held Wednesday, October 4 from 16:20 - 18:30 in the Certification room.

This hands-on session will provide participants with a basic knowledge of Apache Spark. The session has two hands-on labs to teach participants how to start coding using Apache Spark in Jupyter Notebooks on the IBM Data Science Experience/IBM Bluemix cloud platform.

Lab One, “Hello Spark”, shows you the basics to create RDDs, pull in data files, and run map, filter and other basic transformation commands.

Lab Two, “Spark SQL”, shows you the basics to create DataFrames, run Spark SQL, create tables, join tables/DataFrames, and visualize Spark SQL results with matplotlib.