Subject: Re: tuning TB RAID-backed filestore to reduce inode/superblock overhead
To: George Michaelson <firstname.lastname@example.org>
From: Bill Studenmund <email@example.com>
Date: 09/15/2004 17:30:57
Content-Type: text/plain; charset=us-ascii
On Tue, Sep 07, 2004 at 07:50:09AM +0100, David Laight wrote:
> On Tue, Sep 07, 2004 at 11:36:38AM +1000, George Michaelson wrote:
> > Is there a HOWTO for BSD which explains what is 'reasonable' overhead t=
> > towards in constructing a 1.5Tb filestore on RAID?
> > A co-worker just complained the consumed space for 1.5tb looked to be 1=
> > that seems a very high overhead. -no tunefs, no newfs options, no RAID =
> > of the box.
What criteria did the co-worker use to get that 100 GB? Depending on what=
value s/he used, it might not be that far out of line. 100 GB / 1.5 TB =3D=
100 / (1.5 * 1024) =3D=3D 6.5%. We normally reserve 5% of the FS just for r=
(subtracted out of the "Avail" number reported in df), so 6.5% is really=20
> > Any simple guidance would be appreciated.
> The big overhead is the inodes, by default there is one inode for every
> 4 fragments (if not set by -i bytes-per-inode to newfs).
> With 1k fragments this is 1 inode for every 4k, FFSv1 inodes are 128
> bytes - so that is 1/32th of the disk space. FFSv2 inodes are twice
> the size so will use 1/16th of the space.
Just for reference, that's 3.1% or 6.2% overhead for an inode
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)
-----END PGP SIGNATURE-----