From owner-freebsd-arch Fri Jul 7 10:44:30 2000 Delivered-To: freebsd-arch@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by hub.freebsd.org (Postfix) with ESMTP id EADDC37B8E6 for ; Fri, 7 Jul 2000 10:44:26 -0700 (PDT) (envelope-from bright@fw.wintelcom.net) Received: (from bright@localhost) by fw.wintelcom.net (8.10.0/8.10.0) id e67HiMH19071; Fri, 7 Jul 2000 10:44:22 -0700 (PDT) Date: Fri, 7 Jul 2000 10:44:22 -0700 From: Alfred Perlstein To: Marius Bendiksen Cc: Bill Fumerola , freebsd-arch@FreeBSD.ORG Subject: Re: Alterations to vops Message-ID: <20000707104422.P25571@fw.wintelcom.net> References: <20000707080644.H25571@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: ; from mbendiks@eunet.no on Fri, Jul 07, 2000 at 07:42:18PM +0200 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG * Marius Bendiksen [000707 10:42] wrote: > > I think it's AIX that has a tuneable per-process limit on outstanding > > IO. It would be an interesting thing to implement. > > A more complex I/O system, with scheduling and good async operation, > would also be interesting. For example, reads would be balanced more > evenly across tasks, while writes would be buffered for larger chunk > writes. We have that, see vfs_cluster.c, but if we could limit the outstanding IO per process whether read or write we could throttle procs so that they can't saturate the disks. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message