1.1 Resource Requirements and Limits

 R

K8s can throttle the CPU of container so that it cannot exceed the limit. But the container can use more memory than its limit and then K8s will terminate the pod in the case.

1.2 Multiple Scheduler

Deploy additional scheduler:

1. First get the scheduler binary.
wget https://storage.googleapis.com/kubernetes-release/release/v1.12.0/bin/linux/amd64/kube-scheduler
2. Deploy the scheduler in one of the following ways.

3. Specify the scheduler in the pod.

1.2.1 Deploy as service

Creat a service for shedular
my-custom-scheduler.service
  ExecStart=/usr/local/bin/kube-scheduler \\
  --config=/etc/kubernetes/config/kube-scheduler.yaml \\
  my-custom-scheduler.service
  --scheduler-name= my-custom-scheduler

Background: How to deploy scheduler as a service.
https://blog.csdn.net/qq_40460909/article/details/90646787

1.2.2 Deploy with kubeadm (static pods)

etc/kubernetes/my-custom-scheduler.yaml