GDG Cloud Jakarta - GKE 101

Transcript

1. GKE 101 Ananda Dwi Rahmawati ⏰ March, 25 2021
2. Agenda 1. Container and Orchestration 2. Kubernetes Concept & GKE Concept 3. Understanding Clusters and Nodes 4. Pods, Deployments and ReplicaSets 5. Services and Network Ingress 6. Deployment Patterns 7. Stateful Applications and Storage 8. Secrets and Configurations
3. Container and Orchestration
4. Monolithic applications were created from small applications and then built up to create a tiered architecture that separated the frontend from the backend, and the backend from the data sources. Microservices - also known as the microservice architecture - is an architectural style that structures an application as a collection of services that are: • Highly maintainable and testable • Loosely coupled • Independently deployable • Organized around business capabilities • Owned by a small team Reference: https://microservices.io/ Monolithic and Microservices
5. A container image is a lightweight, stand-alone, executable package of a piece of software that includes everything needed to run it: code, runtime, system tools, system libraries, settings. Containerization based on virtualization allows any application bundled in a container which can be run without the hassles of any dependency like libraries. They are a kind of isolated partition inside a single operating system. Container orchestration is a fundamental enabler of digital transformation initiatives. It allows developers and IT organizations to better manage application life cycles. For example: Docker Swarm, Kubernetes, OpenShift, Mesos and Marathon Container and Orchestration
6. Kubernetes & GKE Concept
7. Greek for “pilot” or “Helmsman of a ship” What Does "Kubernetes" Mean?
8. • Project from Google as an open-source container orchestration platform. • Built from the experiences of developing and running Google’s Borg and Omega. • Designed from the ground-up as a loosely coupled collection of components centered around deploying, maintaining, and scaling workloads. What is Kubernetes?
9. • Auto Scaling Easily scale applications up and down manually or based CPU utilization. • Bin Packing Optimize your compute usage by bin packing containers. • Rollout Automation Progressively roll out changes while automatically monitoring applications health. • Self Healing Containers that fail health checks are automatically replaced. • Service Discovery Simple DNS and load balancing provided out of the box. • Storage Automation Easily manage persistent cloud disks or mount other storage providers. Why Kubernetes?
10. Understanding Clusters and Nodes
12. Pods, Deployments and ReplicaSets
13. • A Kubernetes object All configuration in Kubernetes happens through creating objects and object states. • A logical application-centric unit Can comprise one or more containers that share a file system and a network IP address. • The smallest thing you can deploy You can’t deploy containers unless they’re inside a Pod. What is a Pod?
14. • Declare Define the pod in a YAML configuration file. • Apply Apply the configuration by sending it to the master API. • State The state of the cluster is updated to reflect the new declaration. Creating Pods
15. Pods on Nodes GKE Nodes Pod my-web 10.Y.Y.Y 10.X.X.X nginx Container
16. • Represent a set of multiple, identical Pods with no unique identities. • It runs multiple replicas of your application and automatically replaces any instances that fail or become unresponsive. Deployments
17. ReplicaSet Node1 Node2 Node3 ReplicaSet nginx nginx nginx nginx nginx ReplicaSet
18. ReplicaSet Deployment ReplicaSet v2 nginx nginx nginx ReplicaSet v1 nginx nginx nginx
19. Services and Network Ingress
20. 1. Exposes Pods to the network A set of Pods will be given a single fixed IP address. Traffic is round-robined between Pods in the service. 2. Chooses Pods with Selectors Selectors match groups of key-value pairs from Pod metadata. 3. Exposes Pods in different ways Choose from ClusterIP, NodePort, or LoadBalancer. What Does a Service Do?
21. Service Selector Patterns nginx Pod app=nginx env=prod nginx Pod app=nginx env=prod nginx Pod app=nginx env=staging nginx Pod app=nginx env=staging “Web” Service app=nginx
22. Service Selector Patterns nginx Pod app=nginx env=prod nginx Pod app=nginx env=prod nginx Pod app=nginx env=staging nginx Pod app=nginx env=staging Web-Staging Service app=nginx Web-prod Service app=nginx env=prod env=staging web-staging.default.svc.cluster.local
23. Deployment Patterns
24. Rolling Updates ReplicaSet v1 App:v1 App:v1 App:v1 ReplicaSet v2 App:v1 App:v1 App:v1 Services
25. Canary Deployments ReplicaSet app:v2 Service Deployment ReplicaSet app:v1 app:v1 app:v1 app=frontend env=prod Deployment app=frontend env=canary app=frontend
26. Blue/Green Deployments Service Blue Deployment Green Deployment ReplicaSet app:v1 app:v1 app:v1 app=frontend version=1 app=frontend, version=1 ReplicaSet app:v2 app:v2 app:v2 app=frontend version=2
27. Blue/Green Deployments Service Blue Deployment Green Deployment ReplicaSet app:v1 app:v1 app:v1 app=frontend version=1 app=frontend, version=2 ReplicaSet app:v2 app:v2 app:v2 app=frontend version=2
28. Blue/Green Deployments Service Blue Deployment Green Deployment ReplicaSet app:v3 app:v3 app:v3 app=frontend version=3 app=frontend, version=3 ReplicaSet app:v2 app:v2 app:v2 app=frontend version=2
29. Stateful Applications and Storage
30. Local Pods Storage Pod my-app /tmp Pod my-app /tmp Pod my-app /tmp
31. Volumes Pod myapp /mydata
32. • PersistentVolume ◦ Independent lifecycle ◦ Dynamically Provisioned ◦ Storage Class + Capacity • PersistentVolumeClaim ◦ A claim to consume a matching PersistentVolume ◦ With a defined AccessMode • StorageClass ◦ Regional Persistent Disk ◦ SSD Disks Persistent Volumes
33. • Manages Pods using a spec • Maintains an identity per Pod • Guarantees uniqueness and ordering • Ordering applies to deployment, scaling, and updates StatefulSet app-1 app-2
34. • Stateless ◦ Pods added, removed, or restarted at will ◦ Individual Pods have no data to persist or concept of state • Stateful ◦ Pods require an identity ◦ Pods need to store data ◦ Deployment and scaling must be logically managed Stateless vs Stateful Applications
35. Secrets and Configurations
36. Container from immutable image ConfigMaps Pod myapp CAR_MAKE CAR_MODEL
37. ConfigMaps
38. • Data is encoded in etcd • Secrets can be encrypted. • Access to secrets can be restricted using RBAC Secrets
39. • https://github.com/misskecupbung/gke-101 • http://misskecupbung.wordpress.com/ • https://cloud.google.com/kubernetes-engine/docs/concepts/cluster-architecture • https://cloud.google.com/kubernetes-engine/docs/concepts/deployment • https://cloud.google.com/kubernetes-engine/docs/concepts/service • https://cloud.google.com/kubernetes-engine/docs/how-to/updating-apps • https://cloud.google.com/solutions/implementing-deployment-and-testing-strategies-o n-gke • https://cloud.google.com/kubernetes-engine/docs/how-to/stateful-apps • https://cloud.google.com/kubernetes-engine/docs/concepts/storage-overview • https://cloud.google.com/kubernetes-engine/docs/concepts/secret • https://cloud.google.com/kubernetes-engine/docs/concepts/configmap References
40. Thank you!