Skip to content

slurmq

GPU quota management for Slurm clusters.

slurmq helps HPC administrators and users track GPU usage, enforce quotas, and generate reports.

$ slurmq check

╭──────────────────── GPU Quota Report ────────────────────╮
│                                                          │
│   Cluster:  Stella HPC                                   │
│   User:     dedalus                                      │
│   QoS:      medium                                       │
│                                                          │
│   ████████████████████░░░░░░░░░░ 68.5%                   │
│                                                          │
│   Used:      342.5 GPU-hours                             │
│   Remaining: 157.5 GPU-hours                             │
│   Quota:     500 GPU-hours (rolling 30 days)             │
│                                                          │
╰──────────────────────────────────────────────────────────╯

Features

  • Check quota - See your GPU usage at a glance
  • Multi-cluster - Switch between clusters with --cluster
  • Configurable - TOML config files with env var overrides
  • Reports - Generate usage reports in JSON, CSV, or rich tables
  • Monitoring - Live dashboard with optional quota enforcement
  • Analytics - Wait times, utilization stats, month-over-month trends
  • Scriptable - JSON output, quiet mode, exit codes for automation

Quick Start

# Install
uv tool install slurmq

# Initialize config
slurmq config init

# Check your quota
slurmq check

Philosophy

slurmq follows the Unix philosophy:

  • Do one thing well - GPU quota management
  • Composable - Works with jq, cron, scripts
  • Configurable - Defaults that work, overrides when you need them
  • No surprises - Clear output, predictable behavior

Next Steps

For LLMs:

Viewing locally: uv sync --extra docs && uv run mkdocs serve -> localhost:8000