From owner-freebsd-isp Thu Jul 29 2: 2:42 1999 Delivered-To: freebsd-isp@freebsd.org Received: from hack.babel.dk (hack.babel.dk [194.255.106.77]) by hub.freebsd.org (Postfix) with ESMTP id 1466A15520 for ; Thu, 29 Jul 1999 02:02:30 -0700 (PDT) (envelope-from shredder@hack.babel.dk) Received: from localhost (shredder@localhost) by hack.babel.dk (8.9.3/8.9.1) with SMTP id LAA18615 for ; Thu, 29 Jul 1999 11:01:30 +0200 (CEST) Date: Thu, 29 Jul 1999 11:01:29 +0200 (CEST) From: chrw To: freebsd-isp@freebsd.org Subject: FreeBSD diskcache Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-isp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Can someone enlightned me how the diskcache disk i/o performance is achieved on freebsd? It seem, if one runs a find thru all the fs, it will cache some, but not all of the fs. If one mounts the fs with the asynch option, this was supposed to increase performance/use of io disk cache. I tested this, and noticed no difference, the disk cache appear to be of a fixed size, regardless of available (free) memory, and regardless of a(sync) operations. Can someone indepth explain this, in perspective of the linux way, with a dynamic disk cache which inc/decreases the size of the diskcache this way adapting to the current level of free memory, and this way enhancing disk performance, and perhaps introducing problems regarding fs corruption in a crash or heavy load situation (?). Mvh/ Kind Regards, CW To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message