Stress CPU

Attack

Generate CPU load for one or more cores. The resources on a host are shared between containers. Running a resource attack for a container is executed in the containers namespace, but affects the host's resources and all containers running on it.

Targets:

Containers

Install now

Stress CPU

Attack

Targets:

Containers

Install now

Stress CPU

Attack

Targets:

Containers

Install now

Stress CPU

Attack

Targets:

Containers

Install now

Go back to list

Introduction

Generate CPU load for one or more cores

Usage

The resources on a host are shared between containers. Running a resource attack for a container is executed in the containers cgroup, but affects the host's resources and all containers running on it.

Use Cases

Test your application's resilience to high CPU load

Parameters

Parameter	Description	Default
CPU Load	How many CPU should be consumed?	100
Container CPUs	How many workers should be used to stress the CPU?	All Cores
Duration	How long should CPU be consumed?	30s

Frequently Asked Questions (FAQs)

Q: My container has a limit of 500m, set the attack to 100% load but only observe 35-45% CPU usage. Why?

A: stress-ng under the hood generates the load and tries to meet the targeted load, but the Kernel scheduler may assign fewer CPU time slots to the process, resulting in lower CPU usage than desired. For target containers with a higher limit, the desired CPU load is more likely to be met.

Useful Templates (4 of 7)

See all

Stress CPU of Kubernetes DaemonSet

Learn how easy you can stress CPU of an entire Kubernetes DaemonSet.

Stress CPU progressively of Kubernetes DaemonSet

Stress the CPU of a Kubernetes DaemonSet progressively to see at which percentage it will be killed by Kubernetes.

Structure

We start by stressing 50% of the Kubernetes DaemonSet's CPU for 30 seconds. Next, we stepwise stress the CPU by 75%, 90%, and 100% - each for 30 seconds. In between, we have small wait steps to ease analysis in external observability tools for each phase.

Progressive

CPU

DaemonSet

Snippet

Kubernetes

Containers

Stress CPU of Kubernetes Deployment

Learn how easy you can stress CPU of an entire Kubernetes deployment.

Stress CPU progressively of Kubernetes Deployment

Stress the CPU of a Kubernetes Deployment progressively to see at which percentage it will be killed by Kubernetes.

Structure

We start by stressing 50% of the Kubernetes Deployment's CPU for 30 seconds. Next, we stepwise stress the CPU by 75%, 90%, and 100% - each for 30 seconds. In between, we have small wait steps to ease analysis in external observability tools for each phase.