Apache Flink is an open-source framework that supports stream processing and batch processing. It was released in May 2011 by Apache Software Foundation. It’s useful in processing bounded and unbounded data. Flink offers multiple APIs at different levels of abstraction and executes arbitrary dataflow programs in a pipelined manner that enables the execution of batch and stream processing programs.

Project Background

• Framework: Apache Flink 
• Author: Apache Software Foundation
• Released: May 2011
• Type: a framework for batch processing
• License: Apache License 2.0
• Language: Java, Scala
• GitHub: apache/flink

Applications

• Perform computations at an in-memory state
• Graph processing
• Machine Learning
• Complex Event Processing
• Event-driven analytics
• Real-time search index building
• Continuous ETL in e-commerce

Summary

• Operates both batch processing and data streaming programs.
• A very low latency runtime program.
• Runs your applications 24×7
• Programs are automatically compiled and optimized into data flow programs.
• At runtime, native iterative algorithms are used.