Back To Schedule
Olympic - Intermediate

Tuesday May 19th, 16:30-17:15

Cloudflow: Streaming Data Pipelines on Kubernetes Made Easy

About this Session

Streaming applications combine many different capabilities, like Kafka topics, processing jobs, storage, and more to implement the requirements of an application. Developers are often left with the burden of managing the complexity of composing end-to-end applications. In this talk, we introduce Cloudflow (, a new open-source project from Lightbend which aims to simplify the lifecycle of applications on Kubernetes.

For developers, it offers an API that supports the creation of streaming components, a blueprint model to compose them, and tooling for a YAML-free deployment.

On Kubernetes, Cloudflow federates the capabilities of operators to orchestrate the application deployment and all its dependencies, reducing complexity and accelerating their time to market.

Using a live example, we demonstrate how Cloudflow facilitates every step of the application delivery, from development and local testing to full-scale deployment on Kubernetes.


Gerard Maas

Gerard Maas is a Principal Engineer at Lightbend. He is a creative soul with a particular interest in streaming systems. He currently contributes to the engineering of Cloudflow at Lightbend, where he focuses on the integration of Structured Streaming and other stream processing technologies. He is the main author of the O'Reilly book Stream Processing with Apache Spark.