Skip to content

stress-ng — System Stress Testing ​

Tool: stress-ngStatus: Not proven on hardware

NOTE

Status: implemented and unit-tested, not proven on hardware. See the implementation status table. Command examples on this page show the intended workflow; the param block of each script is the authority for parameter names. Not proven on a live cluster.

stress-ng performs CPU, memory, cache, and I/O stress testing for capacity planning and burn-in validation. It includes 370+ stressors and runs on Linux VMs via Ansible deployment.


Run This Test ​

Prerequisites ​

  • Linux VMs deployed on your Azure Local cluster
  • Ansible 2.14+ installed on your management workstation
  • SSH key-based access to Linux VMs
  • Framework prerequisites completed

Step 1 — Install stress-ng on target VMs ​

There is no PowerShell install script. The Ansible playbook installs stress-ng on Linux VMs:

bash
ansible-playbook -i common/ansible/inventory/hosts.yml tools/stress-ng/playbooks/deploy-stress-ng.yml

Step 2 — Run a stress test ​

powershell
.\tools\stress-ng\scripts\Start-StressNgTest.ps1 `
    -ClusterName "hci01.corp.infiniteimprobability.com" `
    -Nodes @("hci01-node1", "hci01-node2") `
    -Profile "cpu-stress" `
    -DurationSeconds 600

Step 3 — Collect results ​

powershell
.\tools\stress-ng\scripts\Collect-StressNgResults.ps1 `
    -ClusterName "hci01.corp.infiniteimprobability.com" `
    -Nodes @("hci01-node1", "hci01-node2") `
    -RunId "<run-id-from-previous-step>"

When to Use stress-ng ​

Scenariostress-ng
CPU thermal and power validation✅
Memory stability under pressure✅
Burn-in testing of new cluster nodes✅
Determine maximum safe CPU utilization✅
Cache and TLB stress testing✅

Key Capabilities ​

  • 370+ stressors across CPU, memory, filesystem, network, and OS categories
  • YAML job files for reproducible configurations
  • Metrics — bogo ops/s, thermal zone readings, power measurement
  • Stressor classes for targeted testing (cpu, memory, io, network, etc.)
  • Verification mode for checking compute accuracy under load

What You Get ​

After a test completes, you will have:

  • Bogo ops/s (bogus operations per second) per stressor per node
  • Thermal readings if available on hardware
  • Aggregated metrics across all nodes
  • Reports in PDF, DOCX, and/or XLSX format

File Locations ​

ComponentPath
Scriptstools/stress-ng/scripts/
Workload Profilestools/stress-ng/config/profiles/
Ansible Playbookstools/stress-ng/playbooks/
Alert Rulestools/stress-ng/monitoring/alert-rules.yml
Report Templatestools/stress-ng/reports/templates/
Teststools/stress-ng/tests/
Logstools/stress-ng/logs/

Deep Dive ​

TopicPage
Ansible deployment detailsInstallation
CPU, memory, and I/O profilesWorkload Profiles
Alert rules during test runsMonitoring
Report formats and templatesReporting
Common errors and fixesTroubleshooting

Automate This Test ​

To run stress-ng on a schedule or as part of a CI/CD workflow, see CI/CD Pipelines. The pipeline calls the same scripts listed above.

Released under the MIT License.