Kubernetes Deployment Metrics with GPU
1,497
Get Dashboard

Created 11/1/2018

Updated 11/4/2018

Revision 2

Description

Monitors Kubernetes deployments in cluster using Prometheus. Shows overall cluster CPU / Memory of deployments, replicas in each deployment. Uses Kube state metrics and cAdvisor metrics.

#Updated to work w/ our metrics

Screenshots

Source Grafana.com

Used Metrics 11

container_accelerator_duty_cycle
container_accelerator_memory_total_bytes
container_accelerator_memory_used_bytes
container_cpu_usage_seconds_total
container_memory_working_set_bytes
container_network_receive_bytes_total
container_network_transmit_bytes_total
kube_deployment_status_replicas
kube_deployment_status_replicas_available
machine_cpu_cores
machine_memory_bytes

Get Dashboard✕

Download

Copy to Clipboard