Xref: news-dnh.mv.net comp.os.msdos.djgpp:2510
Path: news-dnh.mv.net!mv!news.sprintlink.net!newsfeed.internetmci.com!news.mathworks.com!news.kei.com!world!blanket.mitre.org!linus.mitre.org!skidmark.mitre.org!dkb
From: "Daryl K. Baker" <dkb AT mitre DOT org>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Lots of small files in DJGPP waste space & should be chained up!
Date: Mon, 9 Oct 1995 13:52:54 -0400
Organization: The MITRE Corporation
Lines: 24
Nntp-Posting-Host: skidmark.mitre.org
To: djgpp AT sun DOT soe DOT clarkson DOT edu
Dj-Gateway: from newsgroup comp.os.msdos.djgpp

On 9 Oct 1995 a4160023 AT horus DOT sara DOT nl wrote:

> A.Appleyard (A DOT APPLEYARD AT fs2 DOT mt DOT umist DOT ac DOT uk) wrote:
> >  The great size of the full DJGPP puts a big strain on people's hard disk
....
> >department has just got 9 new Pentiums whose hard disk clusters are 32K bytes
> >each!! Each little file must start at the start of a cluster. Thus although
> 
> 
> Personally, I prefer harddisk partitions of 255M bytes.
> This is a reasonable compromise between partition size and slack reduction.
>    

What works really well for me is to use either stacker or drivespace 3
(win95) both internally use small (512???) byte clusters or a RLL type
encoding for the pad at the end of the cluster.  This lets me use the
large drive without the overhead of the large clusters.  -- They have been
several articles in the popular PC mags on how the compression works. 
Stacker is supposed to do better by reducing the cluster overhead to a
couple of bytes. 

Daryl Baker
dkb AT mitre DOT org