Scientific software infrastructure gets very complex, very fast. Your application may include a database, a server side web application, client side web application, load balancer, message queues, databases, or a mix of any of these.
You have an RShiny application, and the server licensing doesn't quite fit your needs.
If you need more flexibility, or prefer to maintain control over your deployment, deploying your RShiny application to AWS is an ideal solution for you. This gives you complete control, including if you need additional databases, message queues, or web applications not written in R.
You have a web application written in Python or Dash that includes additional components, such as a database, message queue, job queue, or reliance upon particular datasets.
Your database, datasets on a file system, or job queue, can all be configured within AWS.