Cluster Monitor

The Cluster Monitor provides a real-time overview of the entire HPC cluster. Data refreshes every 10 seconds.

Keyboard shortcut: gc

Sections

Section	Description
Usage: All Partitions	Aggregated CPU, memory, and disk utilization across the whole cluster
Per-Partition Usage	Individual usage cards for each partition (shown when multiple partitions exist)
Usage: Nodes	Color-coded grid showing per-node CPU, memory, and GPU utilization
CPU Temperature	Color-coded temperature indicators per node (if enabled by admin)
Partitions	SLURM partition table: availability, node counts, job limits
Nodes	Detailed node table: state, CPUs, memory, GPUs, features

If the admin has configured a Dashboard link (e.g., Grafana), a button appears at the top to open it.