优先级表示一个作业相对于其他作业的重要性，Volcano 兼容 Kubernetes 中的 Pod 优先级定义（PriorityClass）。启用该能力后，调度器将优先保障高优先级业务调度。集群资源不足时，调度器主动驱逐低优先级业务，保障高优先级业务可以正常调度。

优先级调度与抢占介绍

用户在集群中运行的业务丰富多样，包括核心业务、非核心业务，在线业务、离线业务等，根据业务的重要程度和 SLA 要求，可以对不同业务类型设置相应的高优先级。比如对核心业务和在线业务设置高优先级，可以保证该类业务优先获取集群资源。当集群资源被非核心业务占用，整体资源不足时，如果有新的核心业务提交部署请求，可以通过抢占的方式驱逐部分非核心业务，释放集群资源用于核心业务的调度运行。

调度类型 说明

基于优先级调度

调度器优先保障高优先级业务运行，但不会主动驱逐已运行的低优先级业务。

基于优先级抢占调度

当集群资源不足时，调度器主动驱逐低优先级业务，保障高优先级业务正常调度。

基于优先级调度需要在调度器的 Plugin 中配置 priority

apiVersion: v1
data:
volcano-scheduler.conf: |
actions: "enqueue, allocate, backfill"
tiers:
- plugins:
  - name: priority
  - name: gang
  - name: conformance
- plugins:
  - name: drf
  - name: predicates
  - name: nodeorder
  - name: binpack
kind: ConfigMap
metadata:
  name: volcano-scheduler-configmap
  namespace: volcano-system

基于优先级调度示例

Volcano 插件内置了三个优先级定义：

apiVersion: scheduling.k8s.io/v1
kind: PriorityClass
metadata:
  name: high-priority
value: 100
globalDefault: false
---
apiVersion: scheduling.k8s.io/v1
kind: PriorityClass
metadata:
  name: med-priority
value: 50
globalDefault: false
---
apiVersion: scheduling.k8s.io/v1
kind: PriorityClass
metadata:
  name: low-priority
value: 10
globalDefault: false

查看优先级信息：

kubectl get PriorityClass

创建高优先级的 Volcano job，占用全部资源：

apiVersion: batch.volcano.sh/v1alpha1
kind: Job
metadata:
  name: priority-high
spec:
  schedulerName: volcano
  minAvailable: 4
  priorityClassName: high-priority
  tasks:
    - replicas: 4
      name: "test"
      template:
        spec:
          containers:
            - image: alpine
              command: ["/bin/sh", "-c", "sleep 1000"]
              imagePullPolicy: IfNotPresent
              name: running
              resources:
                requests:
                  cpu: "1"
          restartPolicy: OnFailure

创建中优先级和低优先级的 Volcano job：

apiVersion: batch.volcano.sh/v1alpha1
kind: Job
metadata:
  name: priority-medium
spec:
  schedulerName: volcano
  minAvailable: 4
  priorityClassName: med-priority
  tasks:
    - replicas: 4
      name: "test"
      template:
        spec:
          containers:
            - image: alpine
              command: ["/bin/sh", "-c", "sleep 1000"]
              imagePullPolicy: IfNotPresent
              name: running
              resources:
                requests:
                  cpu: "1"
          restartPolicy: OnFailure
---
apiVersion: batch.volcano.sh/v1alpha1
kind: Job
metadata:
  name: priority-low
spec:
  schedulerName: volcano
  minAvailable: 4
  priorityClassName: low-priority
  tasks:
    - replicas: 4
      name: "test"
      template:
        spec:
          containers:
            - image: alpine
              command: ["/bin/sh", "-c", "sleep 1000"]
              imagePullPolicy: IfNotPresent
              name: running
              resources:
                requests:
                  cpu: "1"
          restartPolicy: OnFailure

查看 PodGroup 信息：

kubectl get pg

基于优先级抢占调度

启用 preempt

actions: "allocate, backfill, preempt"

  1. 先部署中优先级的 volcano job

  2. 再部署高优先级的 volcnao job

结果如下：

priority