Preliminary study on Kubernetes

One. Container:

1. A container is a way to run one or a set of processes, isolating these processes from other processes on the host

2. Containers are similar to virtual machines, but different from virtual machines

Container virtual machine

Kernel calls the host's kernel per virtual machine one

Start speed seconds-level minute

Resources use only the processes in the container occupy the entire virtual machine

Two. What is Kubernets?

1. is a container organizer, or container manager:

Not only manages container start and stop, but also manages the various resources associated with containers: network, persistent storage volumes

2. Declarative type:

You just have to tell it what you want and it will help you achieve your purpose.

3. From Google's Borg,

Open Source: Developing sources

Open governance: Different from "strict open source", do not want to be controlled by a certain company, led by the code to contribute the most

4. Community:

4.1 Weekly meetings, interested people can participate

4.2 sigs Group: Groups of special interests

Three. Kubernetes at the technical level:

1. is a database (ETCD): Stores our "data of our desired state"

2. Pod: A group of symbiotic Docker, for example: a app+ corresponding monitor + corresponding log ...

App is just one part of the pod

3. Label: corresponds to pod one by one

Pod for version 1

Pod for version 2

Same label for grayscale upgrade

4. Kubectl:rest API Interface

Calling interfaces without modifying the configuration document directly

Four. Kubernetes Workflow

Five. Kubernetes Community Development Direction

Make a stable version: new features and stability trade-offs, based on 1.8 to make a stable version.

Six. How to use kubernetes in the enterprise

1. Managing traffic with a load balancer

2. Use persistent volumes to store stateful application data

3. Open the bottom, there is a certain cost of learning, but many people like

Preliminary study on Kubernetes