Date: Tue, 8 Jun 1999 20:40:27 +0300 From: Vallo Kallaste <vallo@matti.ee> To: Rich Winkel <rich@chumbly.math.missouri.edu> Cc: freebsd-questions@freebsd.org Subject: Re: Freebsd FS i/o blocksize? Message-ID: <19990608204027.A25928@myhakas.matti.ee> In-Reply-To: <9906081215.ZM8112@chumbly.math.missouri.edu>; from Rich Winkel on Tue, Jun 08, 1999 at 12:15:39PM -0500 References: <199906072227.RAA06758@chumbly.math.missouri.edu> <19990608180433.B25205@myhakas.matti.ee> <vallo@matti.ee> <9906081215.ZM8112@chumbly.math.missouri.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jun 08, 1999 at 12:15:39PM -0500, Rich Winkel <rich@chumbly.math.missouri.edu> wrote: > > The maximum is 64KB for one transaction if I can recall some of the > > Greg Lehey's messages. The usual is around 8KB per transaction. I can > > be wrong, I'm not expert by any means. > > Thanks!! So it fluctuates even within a single filesystem? > Or are you talking about swap as well? Better don't ask me, here's something from the vinum(8) manpage: <snip> The FreeBSD block I/O system issues requests of between .5kB and 60 kB; a typical mix is somewhere round 8 kB. You can't stop any striping system from breaking a request into two physical requests, and if you do it wrong it can be broken into several. This will result in a significant drop in performance: the decrease in transfer time per disk is offset by the order of magnitude greater increase in latency. <snip> For swap I don't know, I guess it uses raw disk partition(s) and doesn't have filesystem, althought it must have some system to track the data on the swap partition(s). It's all beyond my knowledge and I'm not interested digging into low-level system internals. Not my way :-| -- Vallo Kallaste vallo@matti.ee To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990608204027.A25928>