Stress testing
lastmod 2025-12-05created 2024-12-054 edits
Stress testing is a deliberate and thorough form of testing used to determine the stability of a given system or entity. It involves testing beyond normal operational capacity, often to a breaking point, in order to observe the results. Reasons can include:
- To determine breaking points or safe usage limits
- To confirm that intended specifications are being met
- To determine modes of failure (how exactly a system fails)
- To test stable operation of a system or part of it outside standard usage
Reliability engineers often test items under expected stress, or even under accelerated stress, to determine the operating life of the item or its modes of failure.
Load testing tools
For a list of load testing software, see Load testing#Load testing tools.
CPU Stress test
-
stress -c $(nproc) -
yes > /dev/null &(run as many as there are cores)
GPU Stress test
- gpu-burn
- Run
python -c "import torch; torch.randn(10000,10000).cuda() @ torch.randn(10000,10000).cuda()"repeatedly.
For monitoring, use nvidia-smi --query-gpu=utilization.gpu --format=csv -l 1.