SpinLoad is a simple web server for generating desired load types on demand. Can be used for testing infrastructure, monitoring applications and
alerts, and autoscaling configurations. You can send a POST
to spinload and request 50% CPU usage for a period of
time by using query parameters.
- Leverage hey to send a series of requests for 5 minutes, each requesting 100% CPU usage to test autoscaling at the application or auto-scale group level.
- Request 25% cpu usage for 20 seconds at a time:
curl http://server:1986/load?cpu=25&time=20s
- CPU - Target CPU percent usage to consume. Leverages a simple infinite loop to consume CPU time and GO channels to stay around desired CPU percent.
- Memory - A target memory usage to consume (WIP)
- DiskIO - Desired Disk IOPs to use (WIP)
- Queue Size - Prometheus formatted metric showing an artificial queue size (WIP)