Flink Forward Berlin Preview: The Apache Flink Ecosystem
Hello, Apache Flink Community! A couple of weeks ago, we wrote a preview post highlighting Flink Forward Berlin sessions about use cases and applications.
Today, we’ll run through sessions that cover the Apache Flink ecosystem. From getting data in and out of Flink applications, to deploying in a wide range of environments, to taking advantage of special-purpose libraries, a robust ecosystem is a crucial component of Flink, and we’re excited to have a range of ecosystem-related talks as this year’s event.
If you like what you see, we’d love for you to join us! You can register for the conference here.
Now onto the sessions.
• Flavio Junqueira of Dell EMC and Stephan Ewen, CTO of data Artisans, will detail Flink’s close integration (and the benefits thereof) with Pravega, Dell EMC’s open source streaming storage offering, which was first announced at this April’s Flink Forward San Francisco.
• Kenneth Knowles and Sourabh Bajaj of Google, along with Aljoscha Krettek of data Artisans, will discuss how to write Apache Beam pipelines in Python(!) and run them on Flink.
• Jörg Schad of Mesosphere and Till Rohrmann of data Artisans will discuss how to run Flink on Apache Mesos and DC/OS, popular resource management frameworks that are used by a large and growing segment of the Flink community.
• Steffen Hausmann of Amazon Web Services will share how to build a real-time data processing pipeline with Apache Flink on AWS, integrating Flink with with Amazon Kinesis and Amazon EMR.
• Docker gets some love in this year’s program, with Dominik Bruhn of Relayr discussing how to automate Flink deployment using Docker and Patrick Lucas of data Artisans sharing community work that helped make Flink a first-class citizen in the container world.
• Andrea Spina and Francesco Frontera of Radicalbit will present Flink-JPMML, a Flink-based open source project that enables fast, dynamic, and distributed machine learning model serving, complete with a demo.
• Rubén Casado Tejedor of Accenture will discuss Flink’s Apache Kudu connector share how Kudu is a good fit for storing the real-time views in a Kappa Architecture (a software architecture pattern that makes use of an immutable, append-only log).
• Piotr Wawrzyniak of Orange Polska S.A. will share how to use Apache Samoa’s machine learning methods with Flink in a use case where Apache Samoa is used as stream mining engine, provided with input data from, and sending predictions to, Apache Flink.
Thanks for tuning in, and we hope to see you in Berlin next month.