Michael Hausenblas

Event-driven architectures (EDAs) offer scalability and promote loose coupling between components of a system, leading to greater agility. However, EDAs bring with them observability challenges.

Designing highly available and resilient applications is often an incremental journey. Collecting data helps you understand where these changes are having a positive impact and where improvements are necessary.

Using observability signals such as logs, metrics, and traces in isolation only gets you so far. While time-based, manual correlation is still widely used, there are better ways to jump between different signal types.

When building and operating cloud-native applications, such as containerized microservices in Kubernetes or serverless applications, you don’t want to proceed by guesswork. In other words, you want to capture and collect all relevant signals (logs, metrics, traces) from your applications and infrastructure so that you can act upon it in a timely manner.

In this session, learn how the Commonwealth Bank of Australia (CommBank) built a platform to run containerized applications in a regulated environment and then replicated it across multiple departments using Amazon EKS, AWS CDK, and GitOps. This session covers how to manage multiple multi-team Amazon EKS clusters across multiple AWS accounts while ensuring compliance and observability...

Hear from a CNCF SIG observability (o11y) member on the state of open-source o11y on AWS. From the Fluent Bit-based FireLens to Prometheus metrics in Amazon CloudWatch to OpenTelemetry, there is a lot to discuss and show in the context of microservices, containers, and serverless compute.