[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: On PATH_MAX
From: |
Jonathan S. Shapiro |
Subject: |
RE: On PATH_MAX |
Date: |
Tue, 08 Nov 2005 14:17:45 -0500 |
On Tue, 2005-11-08 at 12:09 -0700, Christopher Nelson wrote:
> >
> > On Tue, 2005-11-08 at 19:20 +0100, Michal Suchanek wrote:
> >
> > > Even if the compare operation is bound (ie there is a limit on the
> > > length of the filename) I do not see how the EROS solution
> > guaratees
> > > any latency constraint. Searching a directory with 4G
> > files can take
> > > quite long.
> >
> > Not if you choose appropriate data structures for storage.
>
> For example, a balanced binary tree can reduce this to about 32
> operations, max.
Or a properly designed single-probe hash can reduce it to O(1). This is
one of those cases where the binary tree isn't the right choice because
of paging behavior.
- Re: On PATH_MAX, (continued)
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX,
Jonathan S. Shapiro <=
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX, Christopher Nelson, 2005/11/08
- RE: On PATH_MAX, Christopher Nelson, 2005/11/09