The Pragmatic Guide to Crafting Observable Serverless Apps

Crafting reliable serverless apps is notoriously hard. Doing it quickly is even harder. Ensuring deployment and production reliability, moving fast without breaking services, and maintaining stability for consumers is a challenge.

The development, deployment, and production stability of serverless apps has been improved by modern observability techniques and distributed tracing with OpenTelemetry. By using OpenTelemetry with your serverless apps, you can avoid unexpected failures, increase efficiency, and reduce MTTR to keep consumers happy.

In this talk I’ll focus on enabling observability for serverless apps that apply to any FaaS provider like AWS Lambda, Cloudflare Workers, and Vercel Functions. I’ll start with an event-driven microservice app running on Cloudflare Workers which uses a key-value store and serverless database for storage. Then, move on to adding OpenTelemetry instrumentation to help ensure reliable deployments while maintaining stability for your consumers.