From owner-freebsd-current@FreeBSD.ORG Tue Jan 27 13:07:07 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 39DEF16A4CF; Tue, 27 Jan 2004 13:07:07 -0800 (PST) Received: from obsecurity.dyndns.org (adsl-67-119-53-122.dsl.lsan03.pacbell.net [67.119.53.122]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9337F43D66; Tue, 27 Jan 2004 13:06:20 -0800 (PST) (envelope-from kris@obsecurity.org) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id D3B4B66C78; Tue, 27 Jan 2004 13:04:45 -0800 (PST) Date: Tue, 27 Jan 2004 13:04:45 -0800 From: Kris Kennaway To: Bj?rn Gr?nvall Message-ID: <20040127210445.GA68107@xor.obsecurity.org> References: <003c01c3de8d$d569edb0$471b3dd4@dual> <20040127201758.13be4f08.bg@sics.se> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="X1bOJ3K7DJ5YkBrT" Content-Disposition: inline In-Reply-To: <20040127201758.13be4f08.bg@sics.se> User-Agent: Mutt/1.4.1i cc: performance@freebsd.org cc: Robert Watson cc: current@freebsd.org Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 27 Jan 2004 21:07:07 -0000 --X1bOJ3K7DJ5YkBrT Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Jan 27, 2004 at 08:17:58PM +0100, Bj?rn Gr?nvall wrote: > On Sat, 24 Jan 2004 21:14:51 -0500 (EST) > Robert Watson wrote: >=20 > > I haven't done much benchmarking on NFS lately, but something worth > > remembering is that people have spent a lot of time researching and > > optimizing TCP for a variety of connection types, whereas the NFS code > > basically has a static implementation of RPC backoff and flow control t= hat > > hasn't evolved much. >=20 > One reason that FreeBSD users experience poor NFSv3/TCP performance is > that the defaults for rsize and wsize is unusually small, only > 8k. Solaris and HP-UX defaults to 32k for a good reason. I guess TCP > simply needs a little bit more data to chew on to be efficient. >=20 > I tested this on 5.2-CURRENT and found that large file read > performance went up from 56Mbit/s to 80Mbit/s, an improvement by 43%. >=20 > I have written a patch that makes FreeBSD use the same defaults as > Solaris and HP-UX. Note that with NFSv3 there is no risk associated > with specifying to large values for [rw]size. The server automatically > limits these values in the fsinfo rpc. Patch is attached. If no-one picks this up in the next few days, can you please send-pr it so it does not get lost? Kris --X1bOJ3K7DJ5YkBrT Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAFtJtWry0BWjoQKURAkpxAJ43irNJqBozM1E1KwZuoBWxrTVf+gCgkxJA f4nz1j1BYHf53bMw0LzyHCE= =YdgB -----END PGP SIGNATURE----- --X1bOJ3K7DJ5YkBrT--