The implemented gym-hpa is a custom OpenAi Gym environment for the training of Reinforcement Learning (RL) agents for auto-scaling research in the Kubernetes (K8s) platform.

Two environments exist based on the Redis Cluster and Online Boutique applications.

Both RL environments have been designed: actions, observations, reward function.

Please check the run.py file to understand how to run the framework.

To run in the real cluster mode, you should add the token to your cluster here

To run the code, go to the folder policies/run and run:

python run.py

Additional arguments can be passed while running run.py. Please check here run.py.

• Jose Santos

• Tim Wauters

• Bruno Volckaert

• Filip de Turck

If you want to contribute, please contact:

Lead developer: Jose Santos

For questions or support, please use GitHub's issue system.

Copyright (c) 2020 Ghent University and IMEC vzw.

Address: IDLab, Ghent University, iGent Toren, Technologiepark-Zwijnaarde 126 B-9052 Gent, Belgium

Email: [email protected].