Prometheus

devops

This item was not updated in last three versions of the Radar. Should it have appeared in one of the more recent editions, there is a good chance it remains pertinent. However, if the item dates back further, its relevance may have diminished and our current evaluation could vary. Regrettably, our capacity to consistently revisit items from past Radar editions is limited.

Jul 2021

Adopt

Prometheus is an open-source monitoring and alerting system. It was the second project within the CNCF which reached the "graduated" status and has since seen a large rate of adoption across many CNCF projects. It primarily utilizes a pull-based metrics flow through HTTP which allows the easy integration of a variety of application-specific metrics sources. Compared to other monitoring systems it stands out in its simple, still powerful and fully code-based configuration and the equally powerful service discovery mechanism.

Prometheus integrates very well with Grafana which is our tool of choice for dashboard visualization. Through the Prometheus Operator project, the monitoring system can be configured through Kubernetes custom resource definitions. These can be shipped by development teams alongside with their application deployments and allow sharing responsibility for monitoring tasks between operations and engineering teams with a clear interface.

With Cortex and Thanos the Prometheus-ecosystem knows two well-settled solutions for high-availability of the underlying time series database and with Amazon Managed Services for Prometheus there's also a SaaS-Solution available.

We use Prometheus in nearly every project, it's an essential part of our underlying operations and also well understood by many development teams.