In this chapter, we get the readers acquainted with Apache Flink as a candidate for real-time processing. While most of the appendages in terms of data source tailing and sink remain the same, the compute methodology is changed to Flink and the integrations and topology wiring are very different for this technology. Here the reader will understand and implement end-to-end Flink processes to parse, transform, and converge compute on real-time streaming data.
In this chapter, we will look at the following topics:
- Flink architecture and execution engine
- Flink basic components and processes
- Integration of source stream to Flink
- Flink processing and computation
- Flink persistence
- FlinkCEP
- Gelly
- Examples and DIY
- Source to sink: Flink execution
- Executing storm topology on Flink