HTTP Load Testing and Benchmarking Tool

This project provides a general-purpose HTTP load-testing and benchmarking library implemented in Python. The tool allows you to specify an HTTP address, generate requests at a given fixed Queries Per Second (QPS), and reports latency statistics and error rates. It is also Dockerized for easy deployment and execution.

Features

• Specify an HTTP address as input.
• Support for a --qps flag to generate requests at a given fixed QPS.
• Reports latency statistics (average, minimum, and maximum latencies) and error rates.
• Extensible design to add more features.
• Unit tests to verify the functionality.
• Dockerized for easy deployment.

Installation

Prerequisites

• Docker
• Python 3.9 (if running locally without Docker)

Build Docker Image

To build the Docker image, navigate to the project directory and run:

docker build -t load-tester .

Usage

Run the Docker Container

Command-Line Arguments

​ • --url: The URL to be tested (required).

​ • --qps: Queries per second (required).

​ • --duration: Duration of the test in seconds (required).

​ • --method: HTTP method to use (default is GET).

​ • --headers: Optional headers as a JSON string.

​ • --body: Optional request body.

Example

docker run --rm load-tester --url http://example.com --qps 10 --duration 60 --method GET

Development

Local Development

If you want to develop and test the tool locally without Docker, you need to have Python 3.9 installed.

1. Install the required dependencies:

   pip install -r requirements.txt
   

2. Run the tool locally:

   python load_tester.py --url http://example.com --qps 10 --duration 60 --method GET

3. Run the tests locally:

   python -m unittest discover -s .