worked well for App Dev. Now time for AI/ML • But, must ensure data scientist are not hindered by structure Why Containers, Kubernetes & Helm? • Container • Contains everything needed to run your application

Node Components reliably • Canary Rollout • Master & Node Component Versions Management Motivation: Work Order Deployment Worker Order • Upgrade Nodes Versions • Upgrade Node 10.10.10.1 • Upgrade docker Upgrade docker • Upgrade kubelet …. Motivation: Work Order Deployment Disadvantages • Inconsistency • Non-failure-aware • Complicated architecture Work order deployment system can not meet the requirements manage Biz-Cluster master components through Kubernetes resource, such as Deployment, Pod, etc. Work Together Achievement • Anyone can operate and maintenance Kubernetes Cluster • Set up & tear

}} Kris Nova Over engineering the core of Kubernetes kops Kris Nova About Me Kris Nova “I work in the cloud...” Kris Nova In my free time I help run a Kubernetes SIG.. Kris Nova . . t h a “api.nivenly.com” So how does it work ? ? ? Kops 1.4 State Store Unmarshal Tasks Map Compare Reconcile Kops 1.4 tasks := []*Task We have to track our work Where do tasks come from ? ? parse at runtime How do we develop ? ? ? No Yes make go-bindata go build ./kops change work? Kops 1.4 fin Kops 1.4 We were really good at dealing with YAML Kops 1.4 And we had a lot

100 Scale down to 3 How do CronHPA work with HPA 07-28 10:00 07-28 11:00 CronHPA takes over the work HPA takes over the work HPA takes over the work If workload still high load and HPA maxReplicas

have the bandwidth to work on all of them at once. Start with a Bang! Multiple Approaches Kubeadm Improve Project X Focus on Kubernetes only Design better abstractions Work with small number of

the SIG VMware introduction session to learn our mission, recent accomplishments and discuss future work. We will also focus on how new contributors can get involved in the SIG. Kubernetes is in the process Storage Interface (CSI) is a standard API allowing a storage provider to write just one plugin that will work for all major container orchestration systems: Kubernetes, Mesos, Docker and Cloud Foundry. Cluster

with K8s Service How to optimize 02 Comparison with industry Performance measurement 03 04 Future work 05 06 Lessons from eBPF What is K8s Service • It exposes a set of pods via VIP using a load balancer __be32 * to, u32 to_size); pass skb->len to bpf_csum_diff() the verifier reports error Future work • Open source • The modification to IPVS is already open source • https://github.com/Tencent/TencentOS-kernel/

busybox command: - wget - "-O" - "/work-dir/index.html" - http://kubernetes.io volumeMounts: - name: workdir mountPath: "/work-dir" dnsPolicy: Default volumes: - name:

所有云提供商这么玩（例如Google Compute Engine不需要分配 NodePort 来使 LoadBalancer ⼯作，但AWS），但当前API是强制要求的。 Future work（未来的⼯作） 将来，代理策略可能会⽐简单的round-robin均衡策略更加的细致，例如Master选举或分⽚。 Services 也将会拥有“真 正的”负载均衡器，在这种情况下，VIP只会将数据包传输到负载均衡器上。 详细信息，请查看Ingress Controller相关的⽂档。 有关在federated cluster中部署Ingress的详细信息，请参阅federation doc 。 Future Work 各种模式的HTTPS/TLS⽀持（例如：SNI、re-encryption） 通过声明请求IP或主机名 结合L4和L7 Ingress 更多Ingress Controllers 请追踪 Replication Controller或Deployment中的副本数，从⽽将观察到的平均CPU利⽤率与⽤户指定的⽬标相匹配。 How does the Horizontal Pod Autoscaler work?（Horizontal Pod Autoscaler如何⼯作） 28-动态⽔平扩容 130 Horizontal Pod Autoscaler被实现为⼀个控制回路，由Controller Manager的

Example of CRD and Resource Item my-crontab.yaml © 2017 Rancher Labs, Inc. How Does The Controller Work ETCD API Server Kubernetes Core controllers added creating running stoped deleted Resource