@ritazzhang OpenAI Scaling Kubernetes to 2,500 Nodes https://blog.openai.com/scaling-kubernetes-to-2500-nodes/ OpenAI Scaling Kubernetes to 2,500 Nodes https://blog.openai.com/scaling-kubernetes-to-2500-nodes/ Azure Container Registry, gcr, ECR etc.) • Container orchestration • Automating deployment, scaling, and management of containerized applications • Declarative • Can be a mix of GPU or CPU nodes

breed-technologies Right tool for the right job Easy prototyping with php - good performance with GoLang Scaling and orchestration Containerized development and production environment Cloud orchestration with nscd ELB RDS IOPS Pods HPA Evaluation & conclusions Evaluation Kubernetes and docker for fast scaling Low price tag on operations No bottlenecks throughout the season 17% Record breaking Second Second screen interaction 62% Conclusion ● Documentation and procedures ● Prometheus ● Scaling ● Best practice ● Locust Thank you for the attention Any questions? Thomas Hector th@peytz.dk

management system ● Automation ○ Scheduling ○ Deployment ○ Healing ○ Discovery/load balancing ○ Scaling Scheduling Scheduler Scheduler Scheduling Scheduler Scheduling color=yellow Discovery Select Kubernetes! Kubernetes Architecture Operating Kubernetes ● Installation ● Upgrade ● Healing ● Scaling ● Security ● Monitoring ● ... Installation - SSH - Install kubelet - $pkgmanager install kubelet Kubernetes? ● Operational expertise around app management in k8s extends to k8s itself ○ E.g. scaling ● Bootstrapping simplified ● Simply cluster life cycle management ○ E.g. updates ● Upstream improvements

HPA-controller Supporting horizontal scaling of business is one of the essential functions on the cloud, but the HPA Controller of native K8S use only one goroutine to handle the scaling of all businesses in the

of build nodes Node group of user applications Scheduling customization Cluster Resource Auto Scaling kubelet can do image GC DevOps Service DevOps Operator DevOps Operator DevOps Service DevOps push metric to gateway if needed • Cluster autoscaler will add/remove node from build group for scaling • HA is guaranteed by cluster HA, k8s Job controller and cluster autoscaler, can also use container

periodically ConfigMap I read configure file Secret I need confidential data HPA I need auto-scaling Understand Kubernetes in 2 min • kubectl run nginx —image=nginx:1.7.9 —replicas=3 apiVersion: periodically ConfigMap I read configure file Secret I need confidential data HPA I need auto-scaling My Awesome Object I have my own special case Demo • I want to have a Asta Xie object into k8s

PaaS 用户 Kubernetes Cert Manager Ingress Let’s Encrypt Flagger Virtual Service Manual Scaling App CRD HPA Knative Service Cert Canary AutoScaler AutoScaling Route Job Deployment 缺乏交互、复用、可移植能 Image ● Push Docker Image Image Registry Operational Configs (YAML) Revision Controller Scaling Controller Rollout Controller kubectl apply Workload Controller Kubernetes metrics traffic

label selector to provide a LB and Service DNS ReplicaSets A cluster wide Pod manager providing Pod scaling DaemonSets A Pod manager to ensure a Pod is scheduled across a Cluster Node set StatefulSets Replicated Deploys & Configures - CFCR - vSphere - NSX-T Integration - Harbor • Manages Cluster Day 2 - Scaling - Patching - Upgrades - Failures Kubernetes (As a Bosh Release) BOSH (Deploys/Manages VMs) CPI

Operator supports add/removing members to a clustered Operand • Operator enables application-aware scaling of the Operand Monitoring • Operator exposing metrics about its health • Operator exposes health Alerting and Events • Operand sends useful alerts • Custom Resources emit custom events Auto-scaling • Operator scales the Operand up under increased load based on Operand metric • Operator scales

reserved. Amazon Confidential Worker Worker Master Worker Worker Master Auto Scaling group AZ1 Region AZ2 Auto Scaling group CloudWatch Logs Elasticsearc h Kiban a Fluentd DaemonSet Kubectl