Running the Observability As a Service For Your Teams With Thanos

SaaS provides managed functionalities to external customers. However, we see a common trend in the CNCF ecosystem to provide specific features in an API-driven fashion to the internal teams. This includes the Prometheus-based monitoring and observability functionalities. Prometheus is a de facto standard for monitoring static and cloud-native workloads, but similar to Google Borgmon it's built on self-hosting premises. Given the data required for reliable monitoring, analysis, and observability, and its multi-tenant, multi-cluster aspect, creating a single observability team focused on providing tooling to others is very common. Running it as a service is a natural step. In this talk, Ben from AWS and Bartek from Red Hat will introduce you to the Thanos project that brings Prometheus API and storage to a scalable, multi-cluster, multi-tenant level. The audience will learn about the latest features that allow the observability teams to seamlessly deploy Thanos in SaaS mode!