Invitation to the June Meeting of the Big Data User Group

Invitation to the June Meeting of the Big Data User Group 150 150 manu.mukundan

Hot Summer & Big Data: shortly before the big summer break we want to focus on hot topics around Big Data again. On June 30, 2016 at 6:30 p.m. we will therefore welcome you again to the Big Data User Group and look forward to an interesting evening with great presentations. We will meet as usual at the Startup Campus Stuttgart, Teckstra├če 62.

Elephant meets layered architecture (Dr. Valentin Zacharias, Daimler TSS GmbH)
What actually happens when a Big Data project needs to be closely integrated with business applications (and not BI like it stands alone)? What if part of a legacy application needs to be ported to Hadoop? What does it mean to test a Big Data solution? What is the integration environment of a data lake? What does source control mean for a Hadoop application? And what is deployment?

This presentation will describe lessons learned, design patterns, and best practices at the interface between Hadoop and enterprise software landscapes. It looks at both integration with Java enterprise applications and migration of legacy applications “to” Hadoop.

Personal:
Valentin Zacharias works as Senior Data Scientist at Daimler TSS at the interface of Big Data and Advanced Analytics. As an in-house consultant, he and his colleagues support the Daimler Group in a variety of areas in order to create added value from existing data more quickly.

Prior to that, Mr. Zacharias worked as a Big Data Consultant at codecentric and as a manager, consultant and researcher in the areas of artificial intelligence and semantics for the Forschungszentrum Informatik in Karlsruhe. He received his PhD from the Karlsruhe Institute of Technology and studied computer science in Berlin and Massachusetts.

Quo Vadis Multi-Model and Graph – DSE Graph, Graph for Elastic or Neo4J? (Thomas Mann, WidasConcepts GmbH)
In the NoSQL and NewSQL area there is the development in the current year to follow the multi-model approach of a single database. The basic idea is often maintained to include the graph model in addition to the existing data models.

How is this trend to be considered from the point of view of architecture, power of modeling, performance and analytical possibilities in the field of data science? Is a classic graph database like Neo4J still the “best” choice? Or is the approach of an operational database – which also provides full modeling power for analytical purposes – the better way to prevent a proliferation in the big data system landscape? Thomas Mann wants to get to the bottom of these questions in his presentation!

The person:
Thomas Mann is Team Leader and Solution Architect at WidasConcepts in the area of Big Data and Data Science. His focus is on the conception, architecture and implementation of Big Data solutions, taking into account modern Data Science aspects. Current activities and projects focus on the banking sector with projects in the area of Data Lake and fraud detection.