Tests for the Distributed and Parallel Image Processing

The following automation is going to be the one that will be used for testing your final project. This is not a must but highly recommended to test your system with this in order to make sure you have everything in place for the final revision.

The provided step have been tested in the cs-box virtualbox.

Video Utils

The video_utils.py scriipt is a helper for extracting hundreds of images from a video. It will provide a high volume of images that can be processed with your system.

• Install dependencies

sudo pacman -Sy python3 python-virtualenv python-pip --noconfirm
virtualenv .venv
source .venv/bin/activate
pip install -r requirements.txt

• Download a sample video from https://sample-videos.com/

curl -Ok https://sample-videos.com/video123/mp4/720/big_buck_bunny_720p_1mb.mp4

• Run the video_utils.py script

python video_utils.py -action extract big_buck_bunny_720p_1mb.mp4 frames

The script will generate a new directory frames with ~130 images.

Test suite

• Login into your system and save the generated token

curl -u username:password -X POST http://localhost:8080/login

• Get Sytem's status

curl -H "Authorization: Bearer <ACCESS_TOKEN>" http://localhost:8080/status

• Create new workload (save the workload_id)

curl -X POST -H "Authorization: Bearer <ACCESS_TOKEN>" http://localhost:8080/workloads

• Get details about workload

curl -H "Authorization: Bearer <ACCESS_TOKEN>" http://localhost:8080/workloads/<workload_id>

• Upload images

python stress_test.py -action push -workload-id <workload_id> -token <token> -frames-path frames

• Get details about workload

curl -H "Authorization: Bearer <ACCESS_TOKEN>" http://localhost:8080/workloads/<workload_id>

• Download filtered images

python3 stress_test.py -action pull -workload-id <workload_id> -image-type filtered -token <token> -frames-path filtered-frames

• Join filtered images into a new video

python3 video_utils.py -action join filtered.mp4 filtered

• Logout from your system

curl -X DELETE -H "Authorization: Bearer <ACCESS_TOKEN>" http://localhost:8080/logout