Now that we've seen how you can package code and deploy (and a couple of other possible ways to deploy) the package, we'll move on to multiple servers and show how you can orchestrate the deployment of a new version of code to your servers. This time, for the sake of variety, we'll show how you can perform a Tomcat deployment.

In this case, we will run an application, yet again on Amazon EC2 servers, but they could be running in your data center or on any other cloud. So far, we've spoken about how code should look on the box in terms of RPM and how to view database upgrades. Here, we'll show how you can remove servers out of service, upgrade them, and add them back to the mix, and we'll do it for all the servers in the mix, though in this case there are only four of them. This is popularly termed Rolling Deployment. If you only deploy new versions to a certain number of these servers based on either location or some other logic, you could view that as...