AI Tools for Troubleshooting and Optimization
Explore how to leverage AWS AI tools such as CloudWatch investigations and Amazon Q Developer for troubleshooting and optimizing cloud applications. Understand how these tools analyze telemetry data to propose root causes and efficiency improvements. Learn validation methods to confirm AI tool suggestions before applying fixes, ensuring safer deployments and cost savings.
Troubleshooting an application usually means jumping between alarms, metrics, log groups, traces, and deployment history to find which change or resource caused the symptom. Optimization is the same kind of search in the other direction: finding the slow code path, the oversized memory setting, or the idle resource that costs money without adding value. AWS AI tools shorten both searches by scanning the telemetry that already exists and proposing where to look and what to change.
A suggestion from these tools is a hypothesis, not a verdict. The skill the exam tests is using the tools to find candidates quickly and then confirming each one with evidence before changing production. This lesson covers the tools for analyzing errors, the tools for finding optimization opportunities, and the validation habits that connect the two.
The diagram below shows the loop that this lesson follows.
Analyzing errors with CloudWatch investigations
CloudWatch investigations is a generative AI assistant in Amazon CloudWatch that helps respond to an operational issue. After an investigation starts, it scans the account’s telemetry and surfaces ...