Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 11 Apr 2005 20:38:12 -0400
From:      Chuck Swiger <cswiger@mac.com>
To:        NMH <drumslayer2@yahoo.com>
Cc:        questions <freebsd-questions@freebsd.org>
Subject:   Re: Hard drive fullness limits information help request
Message-ID:  <425B1874.7040507@mac.com>
In-Reply-To: <20050411220512.95166.qmail@web41810.mail.yahoo.com>
References:  <20050411220512.95166.qmail@web41810.mail.yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help
NMH wrote:
[ ... ]
>  For shame. A "your question is too dumb to have
> written to our mailing list"? I hope you are not
> trying to represent the great open arms of FreeBSD and
> the questions mailing list. My Question is quite
> appropriate for either list. Nor should someone even
> be given the feeling their questions are too basic to
> bother us with! (unless your aim is to drive people
> away)

If we were using Usenet rather than mailing lists, I would agree with your 
position more-- article cross-posting would be handled without duplication of 
resources, and newsreader software would handle xposting so that people only 
read one copy, even if they are subscribed to both newsgroups.

This is not the case with mailing lists.

The reason why this thread isn't really appropriate to freebsd-hardware is 
because it concerns how the filesystem architecture works in general, on any 
type of device from a hard drive, to a floppy drive, to a USB pen device.

> However, I was hoping someone could point me more
> towards a white paper or some such other information.

Sure.   Try looking at:

    zcat /usr/share/doc/smm/05.fastfs/paper.gz | less
    zcat /usr/share/doc/papers/diskperf.ascii.gz | less

Kirk McKusick also wrote a book on this topic, if you want more information.

However, a point to consider is that if you do some benchmarks for yourself 
using your hardware in the circumstances you care about, you will get better 
numbers for your situation than you do looking at generalizations, averages, 
or specific results from some other hardware being tested in some way that may 
or may not resemble your workload.

-- 
-Chuck



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?425B1874.7040507>