Tracing the Untraceable: Observability in Event-Driven Architectures on AWS

In event-driven systems, understanding how data flows through decoupled services is essential—but not always straightforward. Distributed tracing provides the visibility needed to debug issues, monitor performance, and maintain operational resilience at scale.

In this session, we’ll explore how to implement distributed tracing across event-driven architectures on AWS, including scenarios where events originate from external sources like Apache Kafka. You’ll see how AWS CloudWatch and AWS X-Ray can be used together to trace asynchronous workflows involving Lambda, SNS, SQS, and EventBridge—even when they interact with off-AWS systems.

We’ll walk through real-world examples and share strategies for propagating trace context across service boundaries, asynchronous queues, and hybrid environments. Finally, we’ll compare AWS-native observability tools with Datadog, examining integration options, strengths, and trade-offs for building reliable, distributed systems.