Hello, We are trying to understand a deployment architecture for Shiny and I got this question from tech team. I need help and insight answering that.
They want to understand if there is a way Shiny apps be installed where any heavy lifting of R can happen on a different (bigger) server. This will enable them to install Shiny on a node that is shared with multiple front end applications. They are afraid that if a large data processing request (for R) will choke resources for other apps.
Is there a possibility for such architecture where I can write in the Shiny app (or some configuration parameter), which R server it has to use to process?