Yoda performance tests

This project provides tools for performing performance tests for Yoda.

Installation

1. Clone the repository:

   git clone https://github.com/UtrechtUniversity/yoda-performance-tests.git
   cd yoda-performance-tests

2. Install the required dependencies:

   pip install .

   Or if you prefer uv:

   uv lock
   uv sync

   If you want to develop on these tools, also install the dev extras:

   pip install .[dev]

   Or if you prefer uv:

   uv sync --extra dev

Usage Locust

You can run tests using locust (omit uv run if you installed with pip):

uv run locust -f lf_init.py,lf_irods.py --environment environments/development.json

This will run all Tests (tasks contained in classes derived from locust.User) as tests indefinitely. In this the iRODS tests.

There is also the possibility to run a couple of tests after each other in more staged fashion:

uv run locust -f lf_init.py,lf_irods.py,lf_stages.py --environment environments/development.json

Roadmap

• Setting locust framework
• Exploration of staged runs, first version
• IRODSUser tests blueprint
• Portal tests blueprint
• WebDav tests blueprint
• Parameterization of the User classes in a staged run
• Multi worker runs
• Headless, autostart/autostop runs
• Machine readable reporting

Usage of the performance test scripts

There is also still the possibility to the tests as was initially designed:

yoda-performance-tests -s <sessions>

Arguments

The following command-line arguments can be specified:

Argument	Short Form	Type	Default	Description
--sessions	-s	int[]	10	Number of sessions to open in total (default: 10)
--concurrent-sessions	-c	int	2	Specify the number of concurrent sessions to run (default: 2)
--users	-u	str	users.json	Path to the JSON file containing user credentials (default: users.json)
--verbose	-v	bool	False	Verbose mode - display additional information for troubleshooting
--insecure	-k	bool	False	Disables SSL certificate verification
--graph	-g	bool	False	Generate graph of performance test results

Collaboration

This project is a collaboration between Utrecht University and SURF.

License

This project is licensed under the GPL-v3 license. The full license can be found in LICENSE.