DevConf.CZ 2020

Prometheus has become the de-facto monitoring system for Cloud Native applications. Its powerful data model, operational simplicity and reliability have been key factors in its success. However, Prometheus left some important questions unaddressed: How can deployments be made highly available? How can they be scaled out? What about long term storage?

This talk introduces Thanos, a popular CNCF open-source project that takes Prometheus' strong foundations and extends it into a clustered, yet coordination free, globally scalable metric system. It is already used in production by dozens of companies that want to use Prometheus based metrics in high, multi-cloud scale.

During this talk, the audience will:
* Learn about concepts behind Thanos;
* See a live, actionable walk-through of the seamless transformation of a Prometheus setup into a robust, global and durable monitoring system using Thanos; and
* Learn about potential reference architectures that can be used in production.