Using Locust for Distributed Load Testing: Installation, Script Explanation, and Execution

Locust is an open‑source, Python‑based, scalable, distributed load‑testing framework that is easy to learn and use.

The main idea is to simulate a large number of users accessing your website; each user’s behavior is defined by Python code and can be observed in real time through a web interface.

Locust is fully event‑driven and uses gevent (a coroutine‑based library) instead of callbacks, so each virtual user runs in its own greenlet, allowing thousands of concurrent users on a single machine.

Installation is straightforward: pip install locust . For distributed testing you also need pip install pyzmq .

Locust relies on several libraries: gevent (coroutine networking), flask (lightweight web framework), requests (HTTP client), msgpack‑python (binary serialization), six (Python 2/3 compatibility), and pyzmq (ZeroMQ bindings for distributed mode).

The sample script defines a TaskSet subclass with methods such as index , about , and demo , each decorated with @task(weight) to control execution frequency; a WebsiteUser class sets the target host and wait times.

Execution can be started via the web UI, e.g.: if __name__ == '__main__': import os os.system("locust -f godemo.py --host=http://xx.api.xxxxx.net") or directly from the command line: locust -f godemo.py --host=http://xxx.xxx.xx.net . A no‑web mode is also available with options like --csv , --no-web , -c (users), -r (spawn rate), and -t (run time).

The provided script shows how to connect to a MySQL database using pymysql and sshtunnel , fetch random RFID data, and define three tasks ( getRfid , stock , purchase ) with different weights (1, 3, 7) to simulate realistic usage patterns.

Test results are displayed in the UI, showing the total number of simulated users, hatch rate (users spawned per second), and the ability to start swarming; an example screenshot illustrates a run with 200 users and a spawn rate of 2 users per second.

The article concludes with a QR code invitation for readers to follow and grow together.