📁 Cloud Scheduler AI

Project case study

Cloud Scheduler AI

Reduce queue wait time while placing jobs under CPU, RAM and GPU constraints.

PythonGymnasiumMaskablePPOpytest

The approach

Simulates synthetic jobs and a two-server environment. Compares First Fit, Best Fit, Best Fit RAM and two PPO variants on matched seeds, with action masks for invalid placements.

Measured example
First Fit
14.54
Best Fit
13.30
PPO
13.29
Source: project README. Metrics are specific to the evaluation scenario.

Results and scope

What the project demonstrates

Across 100 synthetic scenarios, mean wait was 14.5425 steps for First Fit and 13.2940 for PPO. PPO did not show a clear advantage over Best Fit methods.