File copying is creation of a new file which has the same content as an existing file.

All operating systems include file copying in the user interface, like cp in Unix and copy in MS-DOS; operating systems with GUIs usually provide copy-and-paste or drag-and-drop methods of file copying.  File managers, too, provide an easy way of copying files.

Internally, however, while some systems have specialized APIs for copying files (like CopyFile and CopyFileEx in Windows API), others (like Unix and MS-DOS) fall back to simply reading the contents of the old file and writing it to the new file. This makes little difference with local files, but provide an interesting situation when both the source and target files are located on a remote file server.  Operating systems with specialized file copying APIs are usually able to tell the server to perform the copying locally, without sending file contents over the network, thus greatly improving performance.  Those systems that have no comparable APIs, however, have to read the file contents over the network, and then to send them back, again over the network.  Sometimes remote file copying is performed with a specialized command, like ncopy in MS-DOS clients for Novell NetWare.

Even more complicated situation arises when one needs to copy files between two remote servers.  The simple way is to read data from one server, and then to write them to the second server.  The tricky way is to tell both servers to exchange data directly (and this is hardly implemented anywhere).