Date: Sun, 27 Jul 2003 13:31:02 -0400 From: Bill Moran <wmoran@potentialtech.com> To: Santos <sansan@cas.port995.com> Cc: freebsd-chat@freebsd.org Subject: Re: Defragment HDD Message-ID: <3F240C56.3050800@potentialtech.com> In-Reply-To: <3F2406AF.5080804@cas.port995.com> References: <Pine.BSF.4.21.0307241548210.73690-100000@lexus.isprime.com> <004501c3521d$e8532c40$3501a8c0@pro.sk> <44brvjhdl4.fsf@be-well.ilk.org> <20030727192744.A5069@welearn.com.au> <3F2406AF.5080804@cas.port995.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Santos wrote: > Sue Blake wrote: > >> On Thu, Jul 24, 2003 at 04:30:47PM -0400, Lowell Gilbert wrote: >> >>> "Peter Rosa" <prosa@pro.sk> writes: >>> >>>> OK, but it is not the "real defragmenting" like Norton Speedisk >>>> or MS Defrag on windoze machines. >>>> Is there anything other ? >>> >>> The term doesn't typically refer to quite the same thing on Unix. No >>> defragmentation program of that type is needed, due to different >>> filesystem >>> internals. See the old (but still useful) /usr/share/doc/smm/05.fastfs >>> for a bit of a better introduction. >> >> You'd be surprised how common this defrag request is... > > Read answer #8 on > http://interviews.slashdot.org/article.pl?sid=03/05/05/1225249&mode=nested&tid=126&tid=185&tid=106&tid=163 > I think the same aplies on filesystem fragmentation too. I concur. I once had a discussion like this: client: "Will this new server defrag automatically or will you have to come in to do it?" me: "No, FFS never gets fragmented?" client: <unbelieving stare> "Then why does Microsoft have to be defragmented?" me: "Because they've foolishly discarded 30 years of filesystem research when they developed the NT filesystem." client: "Why would they do that?" me: <shrug> "maybe they want to make additional money from selling defrag software? Maybe they just don't care enough about their customers to really do a good job?" client: "huh ..." This is a guy that I've had as a client for over 10 years. He trusts me implicitly or the conversation might have turned into an argument. Like Sue's story, he insists that I ensure all the workstations and servers are being defragged regularly, unlike Sue, this client trusted me when I told him how it really was with FreeBSD. But the point I wanted to make is that it hasn't achieved a damn thing. FreeBSD is just TOO good. The FreeBSD print server/firewall/mail server runs week after week with zero problems and allows them to do things that would cost thousands on a MS server. (it's been up 65 days right now, and the last time it was down was because the power went out longer than the UPS could keep it going) The Microsoft server is nothing but trouble (about every three or four months, the backup software will freak out and I'll have to reinstall it. I have no explanation for why this happens, I just know that if I reinstall it, things work smoothly for another three or four months) Because of this, they're constantly aware of the Windows server. But they often forget that the FreeBSD server is even there. I've watched over his shoulder ... when business gets slow he'll start up Diskeeper lite and sit and watch the little red blocks move around and turn into blue blocks. When it's all done, he'll start up some programs and click around muttering things like, "Yeah, that helped ... it's a little faster now ..." He's not the only one I've seen do things like this. Fact is, just like viruses, disk fragmentation is one of the few things that the average computer user actually thinks he understands. And since it seems technical, they feel like their entering the edges of the geek realm when they defrag their hard drives. It gives them some feeling that they actually have some control over their computers. Also, it's been _hammered_ into their heads how necessary it is to defrag hard drives. Anyone who has used a FAT or NTFS filesystem for any length of time without defragging and has seen the tremendous performance gain a defrag can cause is thinking, "I wonder if I need to defrag" any time the computer isn't fast enough. Partially because it's the only thing they know to do in such a circumstance. -- Bill Moran Potential Technologies http://www.potentialtech.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3F240C56.3050800>