Welcome to VMAgent’s documentation!

VMAgent is a platform for exploiting Reinforcement Learning (RL) on Virtual Machine (VM) Scheduling tasks. It is developed by the Multi-Aegnt Artificial Intelligence Lab in East China Normal University and Huawei Cloud. It collects one month real scheduling data from huawei cloud and contains multiple practicle VM scheduling scenarios (such as Fading, Rcovering, etc). These scenarios also correspond to the challanges in the RL. Exploiting the design of RL methods in these secenarios help both the RL and VM scheduling community.

Key Components of VMAgent:

• SchedGym (Simulator): it provides many practical scenarios and flexible configurations to define custom scenarios.

• SchedAgent (Algorithms): it provides many popular RL methods as the baselines.

• SchedVis (Visulization): it provides the visualization of schedlueing dynamics on many metrics.

Contents

Installation

• Installation

SchedGym

• Scenarios
  • Scheduling
  • Cluster
  • Request
  • Clustomize Your Scenario
• Dataset
  • Data Format
  • Notes
  • Statistical Analysis
  • Naive Baselines performance
• GYM
  • State
  • Action
  • Reward
  • Transition Function
  • Interaction Example

SchedAgent

• Framework
  • Controller
  • Learner
  • Components
  • Modules
  • Utils
• DQN
  • Example
• A2C
  • Example
• PPO
  • Example
• SAC
  • Example