Terraform: Up and Running, 3rd Edition

Chapter 8. Production-Grade Terraform Code

Building production-grade infrastructure is difficult. And stressful. And time consuming. By production-grade infrastructure, I mean the kind of infrastructure you’d bet your company on. You’re betting that your infrastructure won’t fall over if traffic goes up, or lose your data if there’s an outage, or allow that data to be compromised when hackers try to break in—and if that bet doesn’t work out, your company might go out of business. That’s what’s at stake when I refer to production-grade infrastructure in this chapter.

I’ve had the opportunity to work with hundreds of companies, and based on all of these experiences, here’s roughly how long you should expect your next production-grade infrastructure project to take:

• If you want to deploy a service fully managed by a third party, such as running MySQL using the AWS Relational Database Service (RDS), you can expect it to take you one to two weeks to get that service ready for production.

• If you want to run your own stateless distributed app, such as a cluster of Node.js apps that don’t store any data locally (e.g., they store all their data in RDS) running on top of an AWS Auto Scaling Group (ASG), that will take roughly twice as long, or about two to four weeks to get ready for production.

• If you want to run your own stateful distributed app, such as an Elasticsearch cluster that runs on top of an ASG and stores data on local disks, that will be another order-of-magnitude increase, ...