In Unix, an application accesses a file by referencing a pathname in a single directory tree which contains all of the names of the files in the file system. In an environment where an application has access to several different file systems, which may be attached at any position of the directory tree, it should be able to ascertain whether a file is physically located on the local system or on a remote server. From an implementation point of view, the physical location of a file is usually the system on whose disk the inode for the file resides.