GS-LinkHQ is a reinforcement learning based agent that generates policy for both computation offloading and distributed caching in the GEdge-Platform. The policy supports horizontal (between cloud edge and neighbor cloud edge) and vertical (between cloud edge and core cloud) collaboration.
- Providing offloading service developers (or service providers) with edge resource allocation policies and capabilities required to implement policies for offloading service development
- Supporting offloading service provisioning: Optimized resource allocation
- Resource allocation policy
- Where to deploy
- How to deploy
- Collaborative deployment or Noncollaborative deployment
- Horizontal, vertical and horizontal + vertical collaboration
- Policy provide point
- Service development point
- When quality of service decreases → When scale out or scale up
- Choosing load balancing method
The information considered to decide the policy is as follows.
- Offloading service definition
- CPU specification
- GPU usage and specification
- Storage/cache specification
- Service properties (network delay sensitive, availability sensitive, etc.)
- Cloud edge (include neighbor edge's information) system resource status information
- CPU usage
- GPU usage
- Storage/cache usage
- Network quality status (latency) (Bet CEs, Bet CE and CC)
- Deployed offloading services quality
GS-LinkHQ consists of three subprojects
simulator: Develop proposed methods first with simulator- Virtual Cluster edge(Single cluster based edge) environment
- Include resource map, network performance map, and offloading service(task) definition
- Use Q-Learning, DQN with intelligent technology
testbed: Develop and validate agents on testbed- K8s based single cluster environment
agent: Reinforcement learning agent developed with the above two projects- Real-world (
GEdge-Platform) environment
- Real-world (
- Docker (or Nvidia Docker)
- Docker compose
git clone https://github.com/gedge-platform/gs-linkhq.gitdocker compose up --build