If you’re getting started with Apache Kafka® and event streaming applications, you’ll be pleased to see the variety of languages available to start interacting with the event streaming platform. It goes way beyond the traditional Java clients to include Scala as well. Unfortunately, there is less beginner content for Scala developers compared to Java. Scala developers might feel a little bit left out, and the idea that the two languages are close enough does not help. …
Kafka Streams is the stream processing library included with Apache Kafka. It enables continuous transformation on events at very low latencies. Among all the possible transformations (filters, map, branch, etc.) joins are special due to their stateful aspect and rely on a concept called co-partitioning. It is documented, but this blog post will introduce an example that brings this idea to life, in a more visual and colourful way!
Two streams are described as co-partitioned if:
Kafka-Streams is the stream processing library included in Apache Kafka, a streaming data platform. Because Kafka-Streams is a simple library, and not a framework, it’s used by applications that can be deployed and run in many ways. This article aims to present a few advantages that come with specific practices like containerization and orchestration, and especially autoscaling. Kafka-Streams is meant to be highly scalable. Let’s explore a way to automatically benefit from this elasticity with Kubernetes.
This article is based on the talk Scale in / Scale out with Kafka-Streams and Kubernetes from Xebicon’18. The repository xke-kingof-scaling contains all the…
Software Engineer @XebiaFr — Confluent CC — @ISEP Alumni — @dataxday organiser — #SPP14