As the growth lead of an open-source APM tool, I keep interacting with developers from companies of all shapes and sizes. I recently talked with a developer from a fintech startup in India. The startup provides a payment processing platform that enables businesses to accept payments from customers worldwide. For them, monitoring is critical, but the dev shared how limited they were when exploring an APM tool for their application.
In this blog, we explain how we enable high availability Prometheus using Cortex and Cassandra. This provides a single pane of view across multiple clusters - which enables visualising all monitoring metrics in one go.
Prometheus is undoubtedly growing as the native monitoring tool for Kubernetes. We have been using Prometheus to collect metrics about our infrastructure for a long, but setting it of the box is still painful
In this blog, we shall send observation frequencies in the bucket intervals chosen and aggregate those at the Prometheus back-end.