Spring cloud dataflow tight coupling

0
votes

I'am using Spring cloud stream with RabbitMQ binder running in spring cloud data flow.

My question is Spring Cloud Dataflow introduces a tight coupling with his runtime as well as with imported bulk applications such as FTP source, JDBC sink ..?

The stream pipeline that runs in the dataflow server as well as the tasks can be run without deploying it in the dataflow server or it require extra work to run on another infrastructure?

Thanks,

1
spring-cloud-streamspring-cloud-dataflow

1 Answers

0
votes

I am not following.

  1. Each application is Spring Boot application they are completely independent and do not require data flow server.
  2. To be more specific each app is spring-cloud-stream app, but still first and for most they are Spring Boot apps and can run independently.
  3. Data Flow server does not run the applications. When we say "deploy to data flow server", that is not your typical deployment. Perhaps we could pick a better term but all that dataflow server does is helps with orchestration, management, monitoring, visualization, versioning etc. In fact you can completely shut down a dataflow server after deploying a data flow stream and it will continue to function.
  4. You can construct the dataflow yourself by manually configuring bunch of spring-cloud-stream apps into a flow, so dataflow servers effectively helps with the process that is otherwise can be achived manually.

So, with that said, what exactly do you mean by tight coupling?