TIME: 4:30-5:15 PM
LOCATION: Conference Room 2A


Monitoring is a key aspect of running containerized applications at scale. Though there are quite a few tools out there, Prometheus, an open source project managed by the Cloud Native Computing Foundation, seems to be the preferred tool of numerous cloud-native infrastructures.

Many Prometheus guides are available explaining how Prometheus can be used to monitor cloud-native applications running in a Kubernetes clusters; how Prometheus may be run in a Kubernetes cluster; and how Prometheus can be setup to monitor a Kubernetes cluster.

At the same time many real-life cloud-native deployment scenarios assume running multiple Kubernetes clusters across different environments – separation due to the nature of the environment (dev, qa, prod, etc), or availability concerns (different regions or availability zones), or hybrid nature of the applications (data center, cloud).

As a Prometheus deployment normally requires significant resources, it may be beneficial in such scenarios to consolidate monitoring infrastructure for multiple Kubernetes clusters and cloud-native applications.

In this session, Serge will showcase how Prometheus can be used to effectively collect metrics from multiple clusters; and share the challenges his team faced and how they overcame them.