Apache Flink has seen incredible growth during the last year, both in development and usage, driven, to a large extent, by the fundamental shift in the enterprise from batch to stream processing. A streaming-first architecture enables continuous processing on data that is continuously produced (as it is in most interesting datasets), enabling real-time decisions but also a simplified architecture that can subsume batch processing. Kostas Tzoumas dives into the benefits of using Flink as the central piece of such architecture. In addition, Kostas covers the latest developments in the project and the future roadmap, such as the ability to query the state in the stream processor, new libraries (e.g., SQL and CEP), dynamic scaling, seamless application and Flink updates, and integration between batch and streaming, which leads to radically simplified architecture and deployment. Kostas concludes with a sample of what production users of Flink are currently achieving with the system.
Kostas Tzoumas is a PMC member of the Apache Flink project and cofounder of data Artisans, the company founded by the original development team that created Flink. Kostas has spoken extensively about Flink, including at Hadoop Summit San Jose 2015.
©2016, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org
Apache Hadoop, Hadoop, Apache Spark, Spark, and Apache are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries, and are used with permission. The Apache Software Foundation has no affiliation with and does not endorse, or review the materials provided at this event, which is managed by O'Reilly Media and/or Cloudera.