Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 15 Jul 1999 02:27:20 +0000 (GMT)
From:      Terry Lambert <tlambert@primenet.com>
To:        davids@webmaster.com (David Schwartz)
Cc:        markov@globalnet.co.uk, regnauld@ftf.net, chat@FreeBSD.ORG
Subject:   Re: Known MMAP() race conditions ... ?
Message-ID:  <199907150227.TAA12176@usr07.primenet.com>
In-Reply-To: <000201bece48$2778d4f0$021d85d1@youwant.to> from "David Schwartz" at Jul 14, 99 03:28:10 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> > I guess it depends how full the partition is. My 2.5GB has >2GB of
> > data on it, not a lot of elbow room for Diskeeper to work with.
> 
> 	My recollection is that Microsoft recommends that you keep your NTFS
> partitions less than half full to 'solve' this problem.

Somewhat like FFS recommends a 15% reserve, and used to compromise
for a little fragmentation by reducing that to 10%.

Effectively, block selection in FFS is a hashing operation,
and as anyone who has read Knuth's "Seminumerical Algorithms: Sorting
and Searching" should be able to tell you, has operations degrade
rapidly, even with a "perfect hash", after 85% fill.

Whoever set the current default to 8% made a similar fragmentation
vs. free reserve trade (I hope it was done while conscious of the
actual trade-off being made, given that "the steady state of disks
is full").


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199907150227.TAA12176>