Data-intensive applications need a sophisticated data architecture more than ever before. Finally, the data world is more versatile and voluminous than ever. In addition, Analytics projects are constantly struggling with the challenge of data acquisition and processing, which is often a brake on agility and motivation at the same time. Therefore, a carefully designed data architecture with deliberately applied design principles quickly pays off. The workshop will discuss the following questions:

- The data-centric enterprise - why a data architecture?

- What distinguishes the data warehouse from Data Lake, which common data architecture approaches are used and what role do they play from the point of view of analytics?

- What data types are there, such as transactional versus machine-generated data such as weblogs and JSON formats, and how does their use differ?

- What are the most common data stores (e. g. relational versus NoSQL) and access mechanisms?

- How are raw data prepared for different use/evaluation types? What is the difference in the processing from historical to real-time data?

- How to secure data (re-)usability for business analysts, BI experts and data scientists?

- What are the challenges of a distributed data architecture?

- What do you need to consider if you want to turn a prototype data pipeline into an operational and scalable application?

The conceptual approaches discussed in the workshop will be illustrated with practical examples.

Short Facts

Design the data architecture according to requirements

Use data type- and task-oriented methods and technologies

Design (re)usable data objects

Being able to implement strategies for solving key data challenges