From owner-freebsd-current@FreeBSD.ORG Sun Nov 4 17:18:24 2007 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id C1E6616A41B; Sun, 4 Nov 2007 17:18:24 +0000 (UTC) (envelope-from peter.schuller@infidyne.com) Received: from smtp.infidyne.com (ds9.infidyne.com [88.80.6.206]) by mx1.freebsd.org (Postfix) with ESMTP id 7BCEE13C4AA; Sun, 4 Nov 2007 17:18:24 +0000 (UTC) (envelope-from peter.schuller@infidyne.com) Received: from c-5416e555.03-51-73746f3.cust.bredbandsbolaget.se (c-5416e555.03-51-73746f3.cust.bredbandsbolaget.se [85.229.22.84]) by smtp.infidyne.com (Postfix) with ESMTP id 912A8CF8C; Sun, 4 Nov 2007 17:40:07 +0100 (CET) From: Peter Schuller To: freebsd-current@freebsd.org Date: Sun, 4 Nov 2007 17:41:34 +0100 User-Agent: KMail/1.9.7 References: <200711021208.25913.Thomas.Sparrevohn@btinternet.com> <200711041302.08877.peter.schuller@infidyne.com> <472DBA71.9070401@FreeBSD.org> In-Reply-To: <472DBA71.9070401@FreeBSD.org> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart6180958.RMZWCyUxTt"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <200711041741.43337.peter.schuller@infidyne.com> Cc: Subject: Re: ZFS Hangs X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 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: Sun, 04 Nov 2007 17:18:24 -0000 --nextPart6180958.RMZWCyUxTt Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline > Well the readahead "issues" are just the zfs design, AFAIK. It does > aggressive prefetching to improve performance on the assumption that you > are running with a reasonably fast storage system that can keep up. On > crappy low end disk hardware (e.g. single ATA disk) this can easily > saturate the disk. Well, regardless of the absolute performance you don't want the read-ahead = to=20 consume to much of the relative bandwidth/seek time available. Personally I didn't experience any prooven issues with prefetch; I disabled= it=20 on general recommendation back in the i386/1gb days and told myself I saw a= n=20 improvement. I never looked into it carefully, other than that I previously= =20 had a general feeling that I saw a lot more I/O on the pool than what was=20 read by applications. My memories are vague. =2D-=20 / Peter Schuller PGP userID: 0xE9758B7D or 'Peter Schuller ' Key retrieval: Send an E-Mail to getpgpkey@scode.org E-Mail: peter.schuller@infidyne.com Web: http://www.scode.org --nextPart6180958.RMZWCyUxTt Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQBHLfZHDNor2+l1i30RArC4AJ4yNkNuK9j0MgkPBsX31DxXt/O8qgCeN7pw r9u+2BVu5FPAKWIYsmvvl8o= =WQN5 -----END PGP SIGNATURE----- --nextPart6180958.RMZWCyUxTt--