Monitoring of batch jobs and VMs running on Google Cloud Platform via HEP Cloud

Monitoring for the HEP Cloud / Google Cloud Platform demonstration at Supercomputing2016. This was a proof-of-concept in elastically expanding scientific computing capability for the CMS experiment onto the GCP Read more at the GCP blog.

The monitoring was based on the Fifemon monitoring system, which uses custom Python probes to query the job and system/daemon status of HTCondor pools, store the metrics in Graphite, and display them in Grafana.

HEP Cloud/GCE Monitoring with Grafana