Skip to main content

Metrics Overview

Gallium collects real-time performance metrics for your deployment and displays them in the Console. Metrics cover both deployment-level resource usage and individual virtual machine performance.

Metric Storage and Retention​

Metrics are collected and stored locally on the deployment. They are not stored in the Gallium Console. All metric data is written to a disk in the System Pool.

By default, metrics are retained for 5 days. Older data is automatically removed to manage disk usage.

Metric Scopes​

Gallium collects metrics at two levels:

  • Deployment metrics track resource usage for the physical servers running your workloads: CPU, memory, disk, and network utilization at the hypervisor level.
  • Virtual machine metrics track the performance of individual VMs: CPU, memory, network I/O, disk I/O, and disk IOPS per virtual machine.

Cluster Storage​

In a cluster deployment, metric data is replicated across three hypervisors on disks in the System Pool. This means metrics remain available even if a single hypervisor goes offline.