There are multiple benefits that teams get when they make their systems observable:

• Enhanced customer experience: There is no doubt that the better you understand your system behaviors, the easier it is to maximize customer benefits from your software applications. Product managers can leverage this data to understand how a particular software application is used and carve out a future strategy for new features and enhancements that would delight their customers.

• Efficient troubleshooting: A good observability strategy leads to reduced investigation time around operational issues and outages. This allows the team to have more time for building differentiating features instead of being pulled into long-lasting investigations.

• Visibility: Observability is not only about fire-fighting operational issues but there are sometimes questions you would like to ask yourself to better understand the consumption patterns of your software application. Where are the majority of my users coming from? How much time do they

typically spend on filling up a particular form? How many clicks does it take for a user to order an item on my website? How many users are passing wrong parameters to my APIs, and so on?

Having established a foundational understanding of what observability is and why it matters, let’s discuss some AWS offerings in this space.

Key AWS offerings for monitoring and observability

AWS offers a variety of native services that enable customers to implement end-to-end observability solutions for their application workloads. All three pillars of observability are well covered. CloudWatch is AWS’s primary observability platform, which is a suite of different solutions that address customer needs around logs and metrics collection. For traces, AWS offers the X-Ray service, which makes it very easy to trace individual requests or workflows in your applications.

Interestingly, there are also quite a lot of customers in the observability space that have been working with open source solutions and would like to extend their existing implementations in the cloud. AWS offers managed solutions for Prometheus and Grafana, which we will cover in a lot more detail during the hands-on exercise, later in this chapter. Furthermore, AWS has been quite active in the Cloud Native Computing Foundation’s (CNCF’s) OpenTelemetry project, also known as Otel forshort. It is a vendor-neutral framework for managing telemetry data. OpenTelemetry introduces the concept of a distribution, which is a customized version of a specific OpenTelemetry component. AWS Distro for OpenTelemetry (ADOT) is an AWS distribution based on the OpenTelemetry project.It provides out-of-the-box capabilities to instrument your applications once and send the metrics and data to multiple AWS services, including Amazon CloudWatch, X-Ray, and Amazon Managed Service for Prometheus.

Before diving into the best practices for a good observability strategy, let’s understand various components of AWS’s observability platform – Amazon CloudWatch.