Date: Thu, 26 Jul 2001 09:24:02 -0300 From: "Daniel C. Sobral" <daniel.sobral@tcoip.com.br> To: Mike Hoskins <mike@adept.org> Cc: Jason Andresen <jandrese@mitre.org>, Jason DiCioccio <geniusj@bluenugget.net>, Mike Tancsa <mike@sentex.net>, Sung Nae Cho <sucho2@quasar.phys.vt.edu>, freebsd-stable@FreeBSD.ORG Subject: Re: Regarding hw.ata.wc="1"...... Message-ID: <3B600BE2.5000706@tcoip.com.br> References: <Pine.BSF.4.21.0107240845420.92951-100000@snafu.adept.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Mike Hoskins wrote: > On Tue, 24 Jul 2001, Jason Andresen wrote: > > >>I think the problem was that FreeBSD started getting killed in various >>benchmarks. Suddenly we have a filesystem that writes slower than >>Solaris and that slows down the entire box. Also, as a desktop OS >>many people noticed that their FreeBSD boxes were writing much slower >>than even the Windows NT boxes. It was really painful. >> > > *sigh* > > I think the problem is people reading (and conducting) benchmarks without > understanding what's being tested and how, etc. > > Unfortuneately then, the reason I chose FreeBSD (server-tuned OS) so > many years ago is no longer valid. Of course, I knew this would > eventually happen in our race to keep up with the ever popular penguin (oh > YAY!). The problem is not the cacheing itself, the problem is that most IDE drives will lie about when the data has actually been written. The point, though, is that if you really care about reliability, you shouldn't be using IDE disks in first place. IDE users are those who prefer a cheaper, less reliable hardware, and the misguided ones. -- Daniel C. Sobral (8-DCS) Daniel.Sobral@tcoip.com.br dcs@newsguy.com dcs@freebsd.org capo@notorious.bsdconspiracy.net All I ask of life is a constant and exaggerated sense of my own importance. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3B600BE2.5000706>