Pivotal has released version 1.5 of Spring Cloud Data Flow, a project for building real-time data processing pipelines, with new features including:
- Improvements to the user interface.
- Updated Spring Cloud Stream Application Starters.
- Improvements in metrics.
- Nested splits for composed tasks.
- Improvements with Kubernetes.
We examine a few of the new features here.
As shown in the diagram below, Spring Cloud Data Flow is built upon a number of Spring projects including Spring Boot, Spring Cloud Stream, and Spring Cloud Task.
Getting Started
This quick start guide demonstrates how to setup Spring Cloud Data Flow on a local machine using the docker-compose utility:
DATAFLOW_VERSION=1.5.1.RELEASE docker-compose up
Once established, the Spring Cloud Data Flow dashboard is accessed via http://localhost:9393/dashboard.
UI Improvements
Improvements to Spring Cloud Data Flow user interface include:
- A more lightweight design.
- An improved
Taskstab to match the style of other tabs. - An improved Stream Builder.
- A new paginator for easier navigation.
The quick start guide also demonstrates how to establish a simple logging stream using the built-in visual editor.
At this point, a stream can be automatically deployed, but the improved Stream Builder includes a property editor to customize a stream before deploying it.
The list of deployed streams sports a new paginator component which can be configured for 20, 30, 50, or 100 items per page.
Stream Application Starters
Stream Application Starters and Task Application Starters, standalone applications powered by Spring Boot and Spring Integration, have recently been updated such that applications can run independently on a number of platforms, produce and consume microservices applications, and provide messaging via Apache Kafka or RabbitMQ.
A special version of Spring Initializr, designed for Spring Cloud-related libraries, has been created.
Notice the differences from the regular Spring Initializr application.
