Kubernetes

CONTAINERS
1. standarized, portable and consistent application environments
  1. Control groups (cgroups)
    1. Hardware resources management
      1. Allow the host to share or limit resources for a process or container
  2. Namespaces
    1. Process, filesystem and networking isolation
      1. Container processes are limited to see only what is in the same namespace
  3. Union filesystems
    1. Containers run from an image (.iso)
Components
1. Most common ones for basic deployments
  1. Node
    1. A physical or virtual machine
    2. Where the service is deployed
  2. Pod
    1. the smallest unit of K8's
    2. Runner environment over the Node
      1. Abstraction over container
      2. Helps so it only interacts with the kubernetes layer
        Can run a variety of container technologies
    3. There's usually one container (appliaction) per pod
    4. Each Pod gets an IP of a K8 virtual network
      1. Gets a new IP each time it restarts
  3. Service
    1. Static IP address
      1. Can be attached to a Pod
        If a Pod dies, the service keeps the IP address
      2. Can be external (targets the node) or internal (targets the pod)
    2. Load Balancer
  4. Ingress
    1. Does forwarding from requests to services
    2. Ensures TLS and DNS
2. Meta-configuration
  1. ConfigMap
    1. External configuration for an application
      1. config services URL, etc.
  2. Secret
    1. Contains meta-variables encoded in a base64 format
      1. variables like credentials
3. Data Storage
  1. Volumes
    1. Data in Pods are volatile
    2. Attaches a physical storage in the hard drive to a Pod
      1. Storage can be done on the local machine or remotely (outisde of K8 cluster)
4. Replica and Schedule Management
  1. Deployment
    1. Blueprints for applications running on pods
      1. Abstraction of Pods
    2. Adjust the number of replicas of a Pod
  2. StatefulSet
    1. Avoid data inconsistencies on pods that need persistant data (like datbases)
      1. Control which pods write on volumes
    2. for STATEFUL applications
Architecture
1. Types of nodes
  1. Master
    1. 4 Processes necessary on Master Node
      1. API Server
        Interaction from user to cluster to deploy applications
        Can be a UI, CLI (kubelet)
        Cluster gateway which validates requests
      2. Scheduler
        Decides which node starts an application Pod
        Schedules based on performance and resource usage on Pods
      3. Controller Manager
        Monitors and detects state changes like Pod failures
        Tells Scheduler to re-schedule downed Pods
      4. etcd
        key value stored of the cluster state
        Logs of cluster activities
        Application data is not stored, only cluster related data
  2. Slave
    1. 3 Processes necessary on each node
      1. Container runtime
        Ex. Docker
      2. Kubelet
        Interacts with the container runtime and the pod
        Assign resources to the container
      3. Kube Proxy
        Forwards requests from services to Pod
        Performance of communication
    2. also called Worker node
    3. Does the application (service) processing
It is a Container Orchestrator
1. Makes microservices more available
  1. Services that are Highly maintainable and testable, loosely coupled, independtly deployable and owned by a small team.
2. Multiple development environment scenarios
  1. Cloud
  2. On premises
  3. Hybrid
3. Management of large scale conteneraized applications
  1. Offering availability, scalability and disaster recovery of services

Next up

Kubernetes

Description

Resource summary

Media attachments

Similar

	Created by Andrés Felipe Velandia Orjuela over 3 years ago