You copy software components from a library and place them into your image using load operations. Since the library stores the compiled representations of these components, as well as their source, loading does not require recompilation.

Each load is atomic. This means that either the entire load succeeds, or the entire load cancels. Non-atomic loads can break in the middle and load only parts of components, resulting in an inconsistent image. Because loads are atomic, there is less chance of your image becoming inconsistent.

The process of loading provides developers with the ability to easily share and reuse code amongst the team. It is a simple task to load software components developed by other members of the team. Because the library keeps all versions of software components, you can load either the newest version or a previous version of a component. You can load individual methods, classes, applications, or configuration maps. This loading ability provides fine-grain control over what new components you wish to install in your image. For more information on loading software components, see Loading components from a library.

You can easily remove components from your image using unload operations. For more information on unload operations, see Unloading components from your image.