Monitoring is one of those tasks with no finish line. There is always something that you could add or improve. Because of that, it is important to prioritize the areas on which you will focus most of your efforts, especially at the beginning, or when new services are released. In addition, different services require different levels of attention. For instance, services such as Lambda, S3, or DynamoDB are considered serverless. AWS takes care of almost everything. You don't need to handle failures, security patches, scaling, high availability, and so on. At the opposite end of the spectrum, you have services such as EC2 where, aside from the hardware itself, you control every aspect of the instance, and, therefore, you need to also invest time and effort in monitoring. Luckily for us, most services that AWS creates have a native integration with CloudWatch. Rather than going down the full list of services we used so far and seeing how monitoring is done, we will look at the different ways to add monitoring to our existing templates.