Locust depends on Redis for managing the job queue and Chrome to execute HTTP requests via Puppeteer.
The suggested setup for the these services is below.
In a local setup (e.g.
locust-cli), it is simplest to run services with Docker Compose and Locust jobs within one or more system processes.
Locust is not limited to running on any one cloud service provider however the selected provider must have a service for serverless functions, hosted Redis or equivalent, and Docker container hosting.
Locust has no persistence capabilities built in. Developers are free to choose any persistence layer they see fit. One use case of the
after hook is to persist collected data to a store.
It is not reccomended to use Locust's Redis to store the output of jobs since it may interfere with the internal operation of Locust.