A certain file system uses “systemwide” names; i.e., once one member of the user community uses a name, that name may not be assigned to new files.Most large file systems, however, require only that names be unique with respect to a given user—two different users may choose the same file name without conflict. Discuss the relative merits of these two schemes, considering both implementation and application issues.

Question

What will be an ideal response?

Answer 1

Certainly, in a large installation requiring unique systemwide names, users trying to
name a file might often find that the system would reject the name. In a really large system, this could go on indefinitely and become a real nuisance.Unique systemwide naming is commonly
used on single-user personal computer systems, but here there is no potential for conflict.
From an implementation standpoint, it is actually more efficient to use local naming,
because the system need not check every new file name against the possibly huge list of existing
file names. Systems that use local names often precede the local file name with the unique
user name (for internal purposes), and this, of course, has the effect of forcing unique global
names; the burden of uniqueness in this scheme is not, however, on the user.