From the course: Apache Flink: Real-Time Data Engineering
Join today to access over 24,200 courses taught by industry experts.
State management in Flink - Flink Tutorial
From the course: Apache Flink: Real-Time Data Engineering
State management in Flink
- [Instructor] Unless your streaming application is basic, application and element state always plays a big role in real time processing. We need state management for various use cases, we may need to refer to previous events and metrics, we want to keep track of active browser or user sessions, we want to track custom counters and make decisions based on them. We may also need to store machine learning models and share them across the application. Flink supports state management in two ways. I supports a keyed state feature where it keeps track of state by specific keys. For example, if we do a KeyBy on a data stream by every user, then Flink keeps track of the state by each user. Each user gets their own copy of the state variable that can then be managed based on the user. Flink supports various data types to store states. They include the simple value state to store a single value, list and map states to store…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.