In Chapter 4, Adding Continuous Integration and Continuous Deployment, in order to illustrate how deploying code works with CodeDeploy, we created a very rudimentary architecture to host our staging and production environments. Each of those environments was composed of only a single EC2 instance.

While this served us well to illustrate our CI/CD pipelines, this approach is of course not recommended when actually deploying an application in production. In this chapter, we will fix that design flow and also address most concerns an engineering organization could have with regard to scaling a typical web application.

We will look at scaling from the angle of performance, where an application gets an increasing amount of traffic, and also for redundancy purposes so that we can handle possible failures in our systems.

Scaling isn't necessarily a DevOps engineering philosophy. The cross-disciplinary nature that scaling an application requires, coupled with the AWS offering in this area, makes it a natural fit for an engineering team focusing on implementing DevOps principles using AWS.

Throughout this chapter, we will discover different tools and strategies to scale our application dynamically.

Typically, services are first implemented as a monolith, meaning that a single application powers the entire service. This approach is usually the most convenient to get an application "out of the door fast." In the first part of the chapter, we will see how to scale these types of architecture.

Next, we will see how to utilize some of the AWS managed services to improve performance and lower the costs of running our service.

In the end, we will see how to break out an application further and through a service-oriented architecture, handle even more traffic. So, in this chapter, we will cover the following topics:

• Scaling a monolithic application
• Improving performance and saving money
• Architecting your application to handle massive amounts of traffic