Date: Wed, 1 May 1996 02:23:59 +0300 (EET DST) From: Heikki Suonsivu <hsu@clinet.fi> To: Peter van Heusden <pvh@leftside.its.uct.ac.za> Cc: freebsd-fs@freebsd.org Subject: Re: Compressing filesystem: Technical issues Message-ID: <199604302323.CAA05011@cantina.clinet.fi> In-Reply-To: Peter van Heusden's message of 26 Apr 1996 04:49:56 %2B0300 References: <Pine.BSD.3.91.960426001058.18458A-100000@leftside>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <Pine.BSD.3.91.960426001058.18458A-100000@leftside> Peter van Heusden <pvh@leftside.its.uct.ac.za> writes:
I'm slowly getting started on the issue of writing a compressing
filesystem for BSD. The situation thus far:
1) I'm thinking of a model much like the Netware 4.x one, where a file is
compressed if it has not been 'touched' (ie. read or written) in a
certain time (e.g. a week). It is then decompressed on being 'touched'.
...
You might want to have a look at
@inproceedings{Burrows:92,
AUTHOR="Michael Burrows and Charles Jerian and Butler Lampson and Timoth
y Mann",
TITLE="On-line Data Compression in a Log-structured File System",
BOOKTITLE="ASPLOS V Proceedings",
ORGANIZATION="ACM",
YEAR=1992,
MONTH="October",
PAGES="2--9",
ABSTRACT = "{We have incorporated on-line data compression into the
low levels of a log-structured file system (Rosenblum's {\it Sprite
LFS}). Each block of data or meta-data is compressed as it is
written to the disk and decompressed as it is read. The
log-strucuring overcomes the problems of allocation and
fragmentation for variable-sized blocks. We observe compression
factors ranging from 1.6 to 2.2, using algorithms running from 1.7
to 0.4 MBytes per second in software on a DECstation 5000/200.
System performance is degraded by a few percent for normal
activities (such as compiling or editing), and as much as a factor
of 1.6 for file system intensive operations (such as copying
multi-megabyte files).
Hardware compression devices mesh well with this design. Chips are
already available that operate at speeds exceeding disk transfer
rates, which indicates that hardware compression would not only
remove the performance degradation we observed, but might well
increase the effective disk transfer rate beyond that obtainable
from a system without compression.}"
}
It looks like implementing this on BSD LFS would not be too hard, as the
idea is trivial and BSD LFS is based on the same idea. I do not know how
stable BSD LFS is, anyone tried it lately ?
--
Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@clinet.fi
mobile +358-40-5519679 work +358-0-4375360 fax -4555276 home -8031121
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199604302323.CAA05011>