Apache Flink

Flink Quick Start Java

We’ll cover installation and setup, and the basics of Flink program structure.

Average Sensor Readings

Our example application ingests a stream of temperature measurements. The program converts the temperatures from Fahrenheit to Celsius and computes the average temperature every 5 seconds for each sensor.

Read-Write Dynamic Table

In the Example we are querying a dynamic table and sending the result of the continuous query to another dynamic table.

Postgres to FlinkTable

In the steps described here we will create a batch transformation that will take information from one source (postgresql jdbc table) and write information to another place – Flink table.

Fraud Detector

Fraud detection is critical for financial institutions, online marketplaces, and e-commerce platforms. A financial institution wants to develop a real-time fraud detection system that analyzes thousands of transactions per second to detect potential fraud. Flink can process each transaction as an individual event with millisecond latency. It doesn’t rely on micro-batching, allowing it to provide real-time results.

Session Window

Description: https://zbciok.github.io/docs/streams/apache-flink/Operators/windowing/#session-window

Sliding Window

Description: https://zbciok.github.io/docs/streams/apache-flink/Operators/windowing#sliding-window

Tumbling Window

Description: https://zbciok.github.io/docs/streams/apache-flink/Operators/windowing#tumbling-windows

Taxi Ride Generator

Example that counts the rides for each driver. Note that this is implicitly keeping state for each driver. This sort of simple, non-windowed aggregation on an unbounded set of keys will use an unbounded amount of state. When this is an issue, look at the SQL/Table API, or ProcessFunction, or state TTL, all of which provide mechanisms for expiring state for stale keys.

Flink Kafka Word Count

This is a re-write of the Apache Flink WordCount example using Kafka connectors.

Stock Price Streams

What we are interested in here is the processing part of the data, where we take the data, do some manipulations, and extract insights from it.

Flink ML (Machine Learning) Examples

Flink ML is a library which provides machine learning (ML) APIs and infrastructures that simplify the building of ML pipelines. Users can implement ML algorithms with the standard ML APIs and further use these infrastructures to build ML pipelines for both training and inference jobs.

Flink Quick Start Java We’ll cover installation and setup, and the basics of Flink program structure.	Average Sensor Readings Our example application ingests a stream of temperature measurements. The program converts the temperatures from Fahrenheit to Celsius and computes the average temperature every 5 seconds for each sensor.
Read-Write Dynamic Table In the Example we are querying a dynamic table and sending the result of the continuous query to another dynamic table.	Postgres to FlinkTable In the steps described here we will create a batch transformation that will take information from one source (postgresql jdbc table) and write information to another place – Flink table.
Fraud Detector Fraud detection is critical for financial institutions, online marketplaces, and e-commerce platforms. A financial institution wants to develop a real-time fraud detection system that analyzes thousands of transactions per second to detect potential fraud. Flink can process each transaction as an individual event with millisecond latency. It doesn’t rely on micro-batching, allowing it to provide real-time results.	Session Window Description: https://zbciok.github.io/docs/streams/apache-flink/Operators/windowing/#session-window
Sliding Window Description: https://zbciok.github.io/docs/streams/apache-flink/Operators/windowing#sliding-window	Tumbling Window Description: https://zbciok.github.io/docs/streams/apache-flink/Operators/windowing#tumbling-windows
Taxi Ride Generator Example that counts the rides for each driver. Note that this is implicitly keeping state for each driver. This sort of simple, non-windowed aggregation on an unbounded set of keys will use an unbounded amount of state. When this is an issue, look at the SQL/Table API, or ProcessFunction, or state TTL, all of which provide mechanisms for expiring state for stale keys.	Flink Kafka Word Count This is a re-write of the Apache Flink WordCount example using Kafka connectors.
Stock Price Streams What we are interested in here is the processing part of the data, where we take the data, do some manipulations, and extract insights from it.	Flink ML (Machine Learning) Examples Flink ML is a library which provides machine learning (ML) APIs and infrastructures that simplify the building of ML pipelines. Users can implement ML algorithms with the standard ML APIs and further use these infrastructures to build ML pipelines for both training and inference jobs.