At the beginning of the book, we chose a virtual machine instance without too much thinking. When designing a production infrastructure, we need to consider the application we are about to deploy and its likely usage over the upcoming months. Ideally, we guess the right instance sizes. But at some point, the application outgrows that infrastructure. That’s okay! If we pay attention to what’s happening on our servers, we can prepare for what startup investors cannot stop talking about. Scaling.We’ll go through considerations and misconceptions, discuss vertical and horizontal scaling, see single server compromises, and introduce load balancers for new deployment strategies. We’ll see how to think of scaling as a process rather than an end goal. Step by step, we’ll learn how to give the application extra room for growth. Measuring the load on system resources will keep our infrastructure in check and give us the necessary insights to scale up and down with confidence.