github.com/ld86/docker@v1.7.1-rc3/docs/terms/layer.md (about)

     1  <!--[metadata]>
     2  +++
     3  draft = true
     4  title = "Layers"
     5  description = "Organizing the Docker Root File System"
     6  keywords = ["containers, lxc, concepts, explanation, image,  container"]
     7  [menu.main]
     8  parent = "mn_use_docker"
     9  +++
    10  <![end-metadata]-->
    11  
    12  # Layers
    13  
    14  ## Introduction
    15  
    16  In a traditional Linux boot, the kernel first mounts the root [*File
    17  System*](/terms/filesystem) as read-only, checks its
    18  integrity, and then switches the whole rootfs volume to read-write mode.
    19  
    20  ## Layer
    21  
    22  When Docker mounts the rootfs, it starts read-only, as in a traditional
    23  Linux boot, but then, instead of changing the file system to read-write
    24  mode, it takes advantage of a [union
    25  mount](http://en.wikipedia.org/wiki/Union_mount) to add a read-write
    26  file system *over* the read-only file system. In fact there may be
    27  multiple read-only file systems stacked on top of each other. We think
    28  of each one of these file systems as a **layer**.
    29  
    30  ![](/terms/images/docker-filesystems-multilayer.png)
    31  
    32  At first, the top read-write layer has nothing in it, but any time a
    33  process creates a file, this happens in the top layer. And if something
    34  needs to update an existing file in a lower layer, then the file gets
    35  copied to the upper layer and changes go into the copy. The version of
    36  the file on the lower layer cannot be seen by the applications anymore,
    37  but it is there, unchanged.
    38  
    39  ## Union File System
    40  
    41  We call the union of the read-write layer and all the read-only layers a
    42  **union file system**.