home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.software-eng:2927 comp.arch.storage:531 comp.unix.internals:1589
- Path: sparky!uunet!mcsun!uknet!keele!csa09
- From: csa09@seq1.keele.ac.uk (Paul Singleton)
- Newsgroups: comp.software-eng,comp.arch.storage,comp.unix.internals
- Subject: Re: Extent-based Filesystems (was: Large Application data sets )
- Message-ID: <3587@keele.keele.ac.uk>
- Date: 23 Jul 92 10:53:40 GMT
- References: <1992Jul22.151535.25117@nuchat.sccsi.com>
- Organization: University of Keele, England
- Lines: 24
-
- From article <1992Jul22.151535.25117@nuchat.sccsi.com>, by steve@nuchat.sccsi.com (Steve Nuchia):
- > In article <1992Jul21.113652.4898@metapro.DIALix.oz.au> bernie@metapro.DIALix.oz.au (Bernd Felsche) writes:
-
- >>The only rational existence I can think of for sparse files
- >>is with memory-mapped (or similar) data files.
-
- > A careful reading of the available historical materials indicates
- > that "holey" files were a serendipitous consequence of design decisions
- > made for other reasons. Personally I think it was unfortunate that
- > this essentially accidental feature became enshrined in the documented
- > semantics of the Unix filesystem.
-
- Surely sparse files are just a space optimisation, a compression, performed
- invisibly by the filesystem? It chooses to represent entire blocks of nuls
- in a compact way. I'm surprised that filesystems don't do other forms of
- compression - there's plenty of scope. If there was a free market in
- filesystem implementations, maybe there would be more development of this
- sort. The holiness of files is _independent_ of their semantics (at least,
- of their system-call program-visible semantics).
- ----
- __ __ Paul Singleton (Mr) JANET: paul@uk.ac.keele.cs
- |__) (__ Computer Science Dept. other: paul@cs.keele.ac.uk
- | . __). Keele University, Newcastle, tel: +44 (0)782 621111 x7355
- Staffs ST5 5BG, ENGLAND fax: +44 (0)782 713082
-