Smart-Kube: Energy-aware and Fair Kubernetes Job Scheduler using Deep Reinforcement Learning

[wes968]

When G(t,x) become small -> cost will become high,it is a little bit strange.
Normally,when G(t,x) become small -> cost will become low.
Have I understood wrongly? Thanks.