Date: Mon, 11 Nov 1996 10:27:28 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: bsdcur@shadows.aeon.net (mika ruohotie) Cc: freebsd-current@freebsd.org Subject: Re: ufs is too slow? Message-ID: <199611111727.KAA18328@phaeton.artisoft.com> In-Reply-To: <199611111435.QAA23398@shadows.aeon.net> from "mika ruohotie" at Nov 11, 96 04:35:53 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> i posted this to usenet too... > > i was wondering if there's any truth here... > > my friend told me that the ufs filesystem is too slow for a high end > nntp server usage... > > the average 7 articles per second is supposed to be too much, and that > using some log type filesystem, like xfs (in sgi) would be better... > > he was saying that ufs cant create those 7 files in the second. > > is that so? VXFS is a log structured FS. UFS (the FFS implementation in SVR4) is within 5% of the speed of VXFS, as long as you use a comparable block size. For SVR4 defaults, this block size should be 8k for UFS (FFS). Most of the directory handling code in VXFS was ripped directly from the SVR4 UFS code. A better soloution is to go to a general soft updates mechanism as the bottom end buffer cache interface for FS's. Many (not all) of the benefits of log structuring have been highly exagerated... file create time is one of them. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611111727.KAA18328>