I was keen on using Red Hat or similar flavoured Linux, because of purpose #4 noted above I already knew a bit about Ubuntu and wanted to expand my knowledge sphere, and Red Hat is the flavour that seems to pop up on corporate and government servers. I used the Elastic IP service (fairly cheap) to allocate a permanent IP address to the instance. I chose Amazon Web Services Elastic Cloud to host the server. If you’re interested in Twitter data but not in setting up a Linux server with R, Shiny and PostgreSQL on it you can skip straight to where I talk about getting the data in and analysing it. The end result looks like this (click on the image to go to the actual working app): In this blog post I’ll be noting down the steps necessary for setting up this server, and illustrating purpose 2 with a Shiny app, hosted on the server and drawing data from a database on the server, which is collecting a round-the-clock random sample of Twitter. I couldn’t tell you which of these are the more important.

to make sure I stay familiar with what’s needed to set up such a system from scratch.

to run my pet HappyRrobot twitterbot, having retired the old physical linux box it used to be hosted from.

so I could do round-the-clock data gathering when necessary.

so I could access a reliable setup from anywhere with a web browser, including on networks without the software I need (which happens a lot in corporate/government environments).

For a variety of reasons I wanted a Linux server in the cloud with data science software on it.

Aws shiny server