HPA官网
从最基本的角度来看，Pod 水平自动扩缩控制器根据当前指标和期望指标来计算扩缩比例。

期望副本数 = ceil[当前副本数 * (当前指标 / 期望指标)]
例如，如果当前指标值为 200m，而期望值为 100m，则副本数将加倍， 因为 200.0 / 100.0 == 2.0 如果当前值为 50m，则副本数将减半， 因为 50.0 / 100.0 == 0.5。如果比率足够接近 1.0（在全局可配置的容差范围内，默认为 0.1）， 则控制平面会跳过扩缩操作。

如果 HorizontalPodAutoscaler 指定的是 targetAverageValue 或 targetAverageUtilization， 那么将会把指定 Pod 度量值的平均值做为 currentMetricValue。

1 单项cpu指标扩容缩容

[root@foundation50 k8s]# scp hpa-example.tar server1:    将下载好的镜像拷贝到server1上
[root@server1 ~]# docker load -i hpa-example.tar   导入镜像
[root@server1 ~]# docker tag mirrorgooglecontainers/hpa-example:latest reg.westos.org/library/hpa-example:latest  改标签
[root@server1 ~]# docker push reg.westos.org/library/hpa-example:latest  上传镜像到仓库
[root@server2 ~]# mkdir hpa
[root@server2 ~]# cd hpa/
[root@server2 hpa]# vim hpa.yaml 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: php-apache
spec:
  selector:
    matchLabels:
      run: php-apache
  replicas: 1
  template:
    metadata:
      labels:
        run: php-apache
    spec:
      containers:
      - name: php-apache
        image: hpa-example
        ports:
        - containerPort: 80
        resources:
          limits:
            cpu: 500m
          requests:
            cpu: 200m
---
apiVersion: v1
kind: Service
metadata:
  name: php-apache
  labels:
    run: php-apache
spec:
  ports:
  - port: 80
  selector:
    run: php-apache
[root@server2 hpa]# kubectl apply -f hpa.yaml   创建
deployment.apps/php-apache created
service/php-apache created
[root@server2 ~]# kubectl get svc  查看svc
NAME         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1        <none>        443/TCP   9d
php-apache   ClusterIP   10.102.198.154   <none>        80/TCP    3m47s   创建成功
[root@server2 ~]# kubectl describe svc php-apache 

[root@server2 ~]# kubectl top pod   查看pod所占用资源
NAME                          CPU(cores)   MEMORY(bytes)   
php-apache-594d676555-x7fnp   1m           5Mi      
[root@server2 ~]# kubectl autoscale deployment php-apache --cpu-percent=50 --min=1 --max=10     cpu期望值到达50%，也就是cpu超过0.1m,启动新的pod，pod数范围为1-10
horizontalpodautoscaler.autoscaling/php-apache autoscaled
[root@server2 ~]# kubectl get hpa  
NAME         REFERENCE               TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
php-apache   Deployment/php-apache   0%/50%    1         10        1          14m      cpu目标是50% ，现在只有1个副本
压测：
[root@server2 ~]# kubectl run demo -it --rm --image=busybox --restart=Never -- /bin/sh -c "while sleep 0.01; do wget -q -O- http://php-apache; done"
[root@server2 ~]# kubectl top pod  查看pod资源，pod不断拉伸
NAME                          CPU(cores)   MEMORY(bytes)   
demo                          12m          0Mi             
php-apache-594d676555-2vxdh   92m          8Mi             
php-apache-594d676555-dxz7z   106m         8Mi             
php-apache-594d676555-jmttp   99m          8Mi             
php-apache-594d676555-lc7dm   96m          8Mi             
php-apache-594d676555-nrq9t   90m          8Mi             
php-apache-594d676555-rklfk   93m          8Mi        

[root@server2 ~]# kubectl get hpa
NAME         REFERENCE               TARGETS   MINPODS   MAXPODS   REPLICAS   AGE
php-apache   Deployment/php-apache   43%/50%   1         10        7          12h           cpu达到43%，扩容了8个pod
停止压测
[root@server2 ~]#  kubectl top pod  查看pod资源，cpu资源已经降下来了
NAME                          CPU(cores)   MEMORY(bytes)   
php-apache-594d676555-2vxdh   1m           8Mi             
php-apache-594d676555-dxz7z   1m           8Mi             
php-apache-594d676555-jmttp   1m           8Mi             
php-apache-594d676555-lc7dm   1m           8Mi             
php-apache-594d676555-nrq9t   1m           8Mi             
php-apache-594d676555-rklfk   1m           8Mi             
php-apache-594d676555-z8mjh   1m           8Mi 
[root@server2 ~]#  kubectl top pod  再次查看pod资源，pod已经缩容了，变成了1个
NAME                          CPU(cores)   MEMORY(bytes)   
php-apache-594d676555-jmttp   1m           8Mi   

2 基于多项指标的自动扩缩

[root@server2 hpa]# vim hpa2.yaml  
apiVersion: autoscaling/v2beta2
kind: HorizontalPodAutoscaler
metadata:
  name: php-apache
spec:
  maxReplicas: 10
  minReplicas: 1
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: php-apache
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        averageUtilization: 60    目标值60%
        type: Utilization
  - type: Resource
    resource:
      name: memory
      target:
        averageValue: 50Mi   内存平均值50Mi
        type: AverageValue
[root@server2 hpa]# kubectl apply -f hpa2.yaml  创建     
[root@server2 hpa]# kubectl get hpa  查看hpa
NAME          REFERENCE                TARGETS                         MINPODS   MAXPODS   REPLICAS   AGE
php-example   Deployment/php-example   <unknown>/50Mi, <unknown>/60%   1         10        0          9m24s
[root@server2 hpa]# kubectl delete hpa php-apache   删除之前的
horizontalpodautoscaler.autoscaling "php-apache" deleted