Apache Flink Summer Camp is a one-time online event featuring two-day courses on Stateful Functions and Flink SQL on May 17-21. The training is a great fit for those who are new to Flink and have knowledge of SQL and/or Python.
Check out the details & schedule below and save your spot!
Stream processing is proving to be an important enabler of new business models, and Apache Flink makes it easier to build and manage stream processing applications. A great way to get started with Flink is by leveraging Flink's support for SQL to build and maintain applications for many of the most popular use cases for stream processing, such as data enrichment and ETL, analytics pipelines, and feature generation for machine learning.
Flink SQL is a standards compliant SQL service that leverages the performance, scalability, and consistency of Apache Flink. It supports both batch and streaming use cases, and is optimized for both of these runtime modes. Using SQL for stream processing, however, poses some interesting challenges (and opportunities!) related to temporal ordering, state management, and materialized view maintenance, and this course will cover what you need to know to take full advantage of features like streaming joins, time-versioned tables, and CDC streams. We won’t waste your time teaching you SQL; instead we will focus on showing you how to leverage what you already know to harness the power of Flink.
Querying Dynamic Tables with SQL
The Training Environment
Queries and Time
Joining Dynamic Tables (including time-versioned tables and external lookup tables)
Built-in and User-defined Functions
Using the Table API
Working knowledge of the commonly used parts of SQL, and computer with a modern browser. The training sessions will be delivered via Zoom, and Slack will be used for support during and between the sessions.
Two three-hour training sessions, plus a one-hour wrap-up session.
Hands-on exercises will be provided, along with a cloud-based Flink SQL service and UI. You will be able to immediately explore the concepts that have been presented without having to install anything. You should plan to spend an hour or two after each training session working with the materials provided.
Serverless technologies, like AWS lambda or KNative, have enabled businesses to deploy simple bits of functionality in an easy and cost-effective way. These tools offer clear benefits such as rapid scalability, scale to zero, and zero downtime upgrades. However, today’s technologies are still limited when it comes to building complex workflows and managing application state.
Stateful Functions, an Apache Flink project, is a platform-independent stateful serverless kit that reduces the complexity of building stateful applications in cloud-native environments. It allows developers to deploy complex stateful application code using modern serverless technologies while retaining the consistency of traditional two-tiered architectures.
This workshop will cover the Stateful Functions API and runtime in enough detail to prepare you to be immediately productive. Participants will see how to build complex applications like model serving systems and ride-hailing services while retaining the operational simplicity of serverless architectures.
The Training Environment
Connecting to External Systems
Integrating with Flink DataStream
Building Custom Extensions
This course will require a computer with at least 8 GB of RAM with Docker and kubectl installed. Participants should additionally be comfortable working with small Python applications. Some Java code examples may be shown for reference but expertise is not required. The training sessions will be delivered via Zoom, and Slack will be used for support during and between the sessions.
As a remote, instructor-led training, this is delivered in two 3-hour sessions, with hands-on exercises assigned as homework to be done outside of class. The course will wrap up on May 21 with a 1-hour session.