GE-Global Scheduler 4.0

For Multiple Edge Clusters and Cloud Cluster

• Can Apply yaml for 3L Clusters (Edge Cluster / Near Edge Cluster / Cloud Cluster)
• Updated Center Management Cluster

Add Gedge Configmap for platform services

• name is gedge-system-scheduler-configmap.yaml
• platform service server ip
• platform service port

Updated GEdge-Scheduler Main Core

• Add New Platform info POD at Center Management Cluster
• Update kafka message module code
• Add New topic for processing Rest API for resources of multiple cluster
• Changed to Run Front Server POD and GEdge Scheduler Policy PODs All at Once

Updated GEdge-Scheduler Source Code for Multiple Users and Workspace, Project

• Add Newly Multiple Users
  • Admin User/Normal User
  • login management
• Workspace is created from Cluster Set ( User Selected Clusters)
• User applitions is seperated by project

Update Version clusters for Developing System

• Set K8S version 1.22.x
• Support contaioner runtime are docker,containerd

Add New GEdge Schedluer Policy

• (G)MostRequestedPriority for 3LT
• (G)LowLatencyPriority for 3LT
• GSetClusters for 3LT
• GSelectCluster for 3LT

Get Start

• MongDB

  • kubectl create -f mongo-statefulset.yaml
  • kubectl create -f headless-service.yaml
  • kubectl expose pod gedge-mongo-0 --port 27017 --target-port 27017 --type LoadBalancer -n gedge-system-scheduler

• Redis

  • kubectl apply -k .
  • kubectl apply -f redis_service_LoadBalancer.yaml

• MetalLB

  • kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/namespace.yaml
  • kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/metallb.yaml
  • kubectl apply -f metallb_configmap.yaml

• Kafka

  • kubectl apply -f kafka.yaml
  • kubectl edit deployment gedge-kafka-server -n gedge-system-scheduler
  • Add env
    • name: KAFKA_HOSTNAME value: xxx.xxx.xxx.xxx (host ip)

• Run Platform Info Server

  • kubectl apply -frbac_gedge_platform_info.yaml

• Run Front Server

  • kubectl apply -f rbac_gedge_gsch_server.yaml

• Run Cluster Agent for Each Cluster(s)

  • kubectl apply -f rbac_gedge_cluster_agent.yaml