Reducing reserved space on large filesystems

[Wayne Sierke]

I was very interested to read the following written by Matthew Seaman in
2004.
http://lists.freebsd.org/pipermail/freebsd-questions/2004-January/033754.html

        The parameters given to newfs(8) don't depend on the size of the
        filesystem, so much as the size of the files you intend to store
        within it, and the sort of directory structure you want to use
        (ie. how many files per directory) Many of the tunables in newfs(8) to
        do with the details of disk geometry don't make sense with modern
        drives and should be ignored.
        
        You should try running 'newfs -N' with various values for the '-g'
        average file size and '-h' average numbers of files per directory
        parameters to see what sort of numbers it spits out.  You don't have
        to be too accurate with those file size and files per directory
        estimates -- order of magnitude is generally good enough.  Unless
        you're going to be storing exceptionally large files (say, typical
        size 20Mb) or you want to have directories with 5,000 or more files in
        them, then just using the newfs(8) default values will work very well.
        
        One thing you can do for any file system over about 256Mb is drop the
        free space reserve ('-m' option in newfs(8), or it can be modified in
        an existing filesystem using tunefs(8)).  1% is more than adequate if
        you're creating a multi-gigabyte filesystem.

I'm especially interested in the comment about the 'free space reserve'
which flies in the face of everything I can recall ever reading that has
always mirrored the warnings in tuning(7) and tunefs(8) about the perils
of reducing the reserved space below the default. However I didn't see
any reply to Matthew's email to repudiate his statements.

What are people's experiences in the field? Are the cautions now much
less relevant with modern hard-drive capacities and performance?