From owner-freebsd-fs Tue Feb 27 21:41:16 1996 Return-Path: owner-fs Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA12990 for fs-outgoing; Tue, 27 Feb 1996 21:41:16 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [192.216.222.3]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA12985 for ; Tue, 27 Feb 1996 21:41:13 -0800 (PST) Received: from crh.cl.msu.edu ([35.8.1.24]) by who.cdrom.com (8.6.12/8.6.11) with ESMTP id VAA22275 for ; Tue, 27 Feb 1996 21:41:13 -0800 Received: (from henrich@localhost) by crh.cl.msu.edu (8.6.12/8.6.12) id AAA02027; Wed, 28 Feb 1996 00:39:50 -0500 Date: Wed, 28 Feb 1996 00:39:50 -0500 From: Charles Henrich Message-Id: <199602280539.AAA02027@crh.cl.msu.edu> To: atf3r@stretch.cs.virginia.edu, freebsd-fs@freebsd.org Subject: Re: Compressing filesystem for FreeBSD Newsgroups: lists.freebsd.fs References: <4h0fhq$c40@msunews.cl.msu.edu> X-Newsreader: NN version 6.5.0 #3 (NOV) Sender: owner-fs@freebsd.org Precedence: bulk In lists.freebsd.fs you write: > The real problem with compressing a file system that is often >forgotten is that you are increasing the cost in CPU cycles for a disk >I/O. While I have had dos/linux enthusiasts go on about how compressing >your filesystem can both improve I/O bandwidth and increase available disk >space, the cost is often fogotten. On a multi-user system, the CPU is not >idle during disk I/O. I think there would also be problems with paging >and reads in general. Which block of the disk contains the nth byte? In >short, I don't see it being worth the effort. In real life however, huge numbers of peoples CPU's are idle 99% of the time. I know for one that my P100 is idling along doing not a damn thing for over 50% of the time. If I can use that CPU for something productive, Im all for it. -Crh -- Charles Henrich Michigan State University henrich@msu.edu http://pilot.msu.edu/~henrich