Stream & Batch Processing with Apache Flink and Event-Time Windowing

I'm very pleased have my Friend Robert from DataArtisans in Berlin to start the Munich-based Apache Flink Meetup group.

Please consider joining there as well:

https://www.meetup.com/Apache-Flink-Meetup-Munich/events/226534681/

For the initial event, we have two very interesting talks.

We're pleased to welcome Robert of data Artisans, an avid Flink developer, who will introduce you to Stream & Batch Processing with Apache Flink. The second talk will be held by Matthias J. Sax of Humboldt University Berlin, who will dive deeper into the bowels of processing windows on data streams with Flink.

Preliminary Schedule

• 18.30 – Talk 1 (Robert): Stream & Batch Processing with Apache Flink

• 19.30 – Dinner Time & socialising

• 20.30 – Talk 2 (Matthias): Feeding a Squirrel in Time -- Windows in Apache Flink

As always, this Meetup is free for all attendees & Food and Beverages are provided by our host Inovex GmbH

Abstracts

Stream & Batch Processing with Apache Flink

Apache Flink is an open-source framework for distributed data analysis. Flink is one of the most active big data projects in the Apache Software Foundation and has more than 100 contributors. The core of Flink is a distributed stream processing engine that enables low-latency and high throughput data processing. Flink has high-level APIs for both stream and batch processing.
In the first part of this talk we will briefly look at the history of Flink and then present the APIs and different use cases. Here we will also see how it can be deployed in practice.
In the second part, we will look at the streaming execution engine that powers Flink. Here we will see what makes it tick and also what distinguishes it from other approaches, such as the mini-batch execution model. We will cover Flink's architecture and design decisions that result in Flink's unique set of features.

Feeding a Squirrel in Time -- Windows in Apache Flink

In data stream processing, the dimension of time is a crucial aspect with regard to runtime effects as well as query semantics. There are two different time concepts: event-time also called occurrence time, ie, the time in which an event happens in the real world, and processing-time, ie, the time in which an events gets processed by the system. Both time dimensions deliver different semantics if used to specify windows on a data stream.
In this talk, we give an introduction to time-base data stream
processing and cover Flink's approach to handle windowing. Flink offers a broad range of (non-/)deterministic count-, time-, and custom-windows while offering a simple API to express those.

Please note that this event has also been posted here: https://www.meetup.com/inovex-munich/events/226206381/

We would like to thank our host Inovex GmbH for providing the location, food and drinks.