Containers

Friday, August 9, 2019 11:39 AM

Hypervisor 1 - runs directly on the OS machine hardware. e.g ESX and Zen
Hypervisor 2 - Windows Virtualization; Vmware workstation, virtual box.

Container engines can be freebsd jails, solaris zones, rocket or docker.

Containers have significant smaller overheads and they share the kernel with the host OS; hence they are extremely fast.
Typically a container contains a single process and its libraries and the startup time is dependent on the how long the process takes to start.

Docker containers must run linux distributions and cannot use windows or unix processes.
Docker provides simple command line tools via http APIs for managing and automating the containers applications.
Docker containers ship with containerized applications with its dependencies baked in; hence deployment becomes easier regardless of linux distribution.

The containers that Docker builds are isolated with respect to eight aspects. Part 1
of this book covers each of these aspects through an exploration of Docker container
features. The specific aspects are as follows:
■ PID namespace—Process identifiers and capabilities
■ UTS namespace—Host and domain name
■ MNT namespace—File system access and structure
■ IPC namespace—Process communication over shared memory
■ NET namespace—Network access and structure
■ USR namespace—User names and identifiers
■ chroot()—Controls the location of the file system root
■ cgroups—Resource protection

Linux namespaces and cgroups take care of containers at runtime.

You can think of a Docker container as a physical shipping container. It’s a box where you store and run an application and all of its dependencies.

A Docker image is a bundled snapshot of all the files that should be available to a program running inside a container. You can create as many containers from an
image as you want. But when you do, containers that were started from the same image don’t share changes to their file system.

Training Page 1
Training Page 2