TensorFlow和Kubernetes集群擴容
莫問出處 莫問歸途
本文演示在Kubernetes和TensorFlow集群擴容。
集群如下:(node1已修改,把TA加入node0、node2集群)
一、Docker安裝
位置:node1
安裝:yum install docker
啟動:
systemctl enable docker
systemctl start docker
驗證:
二、ETCD擴容
位置:node1
安裝:yum install etcd
配置:/etc/etcd/etcd.conf
入群:
curl http://192.168.0.180:2379/v2/members -XPOST -H "Content-Type: application/json" -d '{"peerURLs": ["http://192.168.0.181:2380"]}'
啟動:
systemctl enable etcd
systemctl start etcd
驗證:
三、K8S擴容
位置:node1
安裝:
yum install kubernetes-master
yum install kubernetes-node
配置:
/etc/kubernetes/config:
/etc/kubernetes/apiserver:
/etc/kubernetes/controller-manager:
啟動:
systemctl enable kube-apiserver
systemctl start kube-apiserver
systemctl enable kube-controller-manager
systemctl start kube-controller-manager
systemctl enable kube-scheduler
systemctl start kube-scheduler
systemctl enable kubelet
systemctl start kubelet
systemctl enable kube-proxy
systemctl start kube-proxy
驗證:
四、TF擴容
位置:node1
1、提前下載tf鏡像:docker pull tensorflow/tensorflow:latest
備註:docker官方下載速度可能很慢,可以從其他源下載。
2、提前安裝rhsm,否則無法下載pod-infrastructure
yum install *rhsm*
3、RC查看:
kubectl get rc
kubectl get pod
4、RC擴容:2 pod -> 3 pod
5、RC驗證:
6、Service驗證:
kubectl get svc
curl -L http://192.168.0.181:30000
五、使用TF
1、流覽器打開(通過cat /var/log/messages | grep 8888可以找到token):
2、編寫TensorFlow程式,執行"Run":