protocol: TCP $ kubectl get deployment NAME REVISION AGE frontend-c8bb659c5 1 2d15h $ kubectl get components NAME WORKLOAD frontend deployment.apps.k8s.io Component：应用中的一个组成部分，例如容器、 Function或者云服务等 apiVersion: core.oam.dev/v1alpha2 kind: ApplicationConfiguration metadata: name: helloworld spec: components: # 1st component - componentName: frontend traits: - trait: apiVersion: autoscaling/v2beta2 Configuration Application Configuration PaaS A PaaS B Serverless C Components Traits Components Traits Components Traits Components Developers Operators 案例：今天的阿里巴巴应用管理 • 统一、标准 • 用户友好 • 高可扩展 • 全局拉通共建

● A self-driving approach Motivation: microservices ● Increased operational cost ○ a lot of components ○ dynamic dependencies ○ fast deployment iteration ● Solution: automation Cluster management - SSH - Upgrade container runtime - Upgrade Kubelet Upgrade - master - SSH - Upgrade master components Upgrade - etcd - SSH - Upgrade etcd Upgrade kops AWS, GCP API node1 node2 node3 Rollback Self-hosted Kubernetes? What is self-hosted Kubernetes? ● Kubernetes manages own core components ● Core components deployed as native API objects Self-hosted k8s Architecture Why Self-host Kubernetes

& Teardown Cluster fast and convenient • Add & delete Node at any time • Upgrade Master & Node Components reliably • Canary Rollout • Master & Node Component Versions Management Motivation: Work Order Biz-Cluster: used to deploy our application. • Meta-Cluster: used to set up Biz-Cluster master components. We add Biz-Cluster master nodes to Meta-Cluster. • User: SREs who can setup & upgrade Biz-Cluster Kube-on-Kube-Operator: difference Biz-Cluster CRs and Biz-Cluster master components state, and manage Biz-Cluster master components through Kubernetes resource, such as Deployment, Pod, etc. Work Together

component that leads delivery of the pod to failure Is the cluster healthy 1 Are all software components working fine 2 How many failures occurred on the cluster SLIs on Large k8s Cluster 1. Cluster consequently. 4. Centralized Components Availability A ratio value indicates the time in which the cluster is available. It is used to evaluate the master components. The success standard and reason

be/PH-2FfFD2PU 14 VM VM VM VM Kubernetes Trend Worker Node The Kubernetes Master Node Basic Components Master Node ETCD kube-apiserver kube-controller-manager kube-scheduler • Key/Value Store performance and capacity • Affinity/Anti-Affinity Capable The Kubernetes Worker Node Basic Components Master Node ETCD kube-apiserver kube-controller-manager kube-Scheduler Worker Node CRI-containerd Creates virtual IP for external access • Interfaces with local iptables The Kubernetes Runtime Components • Runtime Objects Component Description Pods A grouping of one or more containers as an atomic

•Flavor •ComputeNode Configuration •Kernel params •Environment config •Network Kubernetes •Core components •Addon •Taint Operations Our thinking of datacenter modeling by extending Kubernetes Onboard SaltMaster Step 3. Relax and have a cup of coffee What if salt master down? Upgrade Kubernetes core components Upgrade addons How to upgrade a cluster? ● Kubernetes is amazing on its simple architecture

Inception v3 and transfer learning • Automate repeatable ML experiments with containers • Deploy ML components to Kubernetes with Kubeflow • Scale and test ML experiments with Helm • Manage training jobs and Jupyter notebooks • Model serving – serve exported models with TF Serving or Seldon • Additional components for storage, workflow, etc. Artificial Intelligence solves critical life problems Demo: Find

OpenWhisk Deploy Apache OpenWhisk on Kubernetes • The architecture diagram of OpenWhisk components on Kubernetes, e.g. • https://github.com/apache/incubator-openwhisk-deploy-kube Technical running task, can be exposed as K8s services • In OpenWhisk, usually, we deploy those core components’ dependencies lib or tools as Deployment: – CouchDB – Redis – Zookeeper – Nginx

Kubernetes https://github.com/kubevirt KubeVirt Architecture KubeVirt Application Layout KubeVirt Components • virt-controller • virt-handler • libvirtd KubeVirt Managed Pods • VMI Foo • VMI Bar KubeVirt

使得一个节点无法工作（Pod Failure） ● 观察和检验 ○ QPS 下降之后却再也没有恢复到实验前的水平 ○ 我们找到了一个 Bug ！ 以 FUXI-Lab 为例 ● Testing components(redis rabbitmq scheduler) ● Testing bugs 3+ 20 Welcome to a “bug-free” world! Without worrying