From owner-freebsd-arch@FreeBSD.ORG Mon Aug 4 17:55:29 2003 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D38D437B401; Mon, 4 Aug 2003 17:55:29 -0700 (PDT) Received: from wantadilla.lemis.com (wantadilla.lemis.com [192.109.197.80]) by mx1.FreeBSD.org (Postfix) with ESMTP id E986843FB1; Mon, 4 Aug 2003 17:55:27 -0700 (PDT) (envelope-from grog@lemis.com) Received: by wantadilla.lemis.com (Postfix, from userid 1004) id E3CD1526A1; Tue, 5 Aug 2003 10:25:24 +0930 (CST) Date: Tue, 5 Aug 2003 10:25:24 +0930 From: Greg 'groggy' Lehey To: Poul-Henning Kamp Message-ID: <20030805005524.GS95375@wantadilla.lemis.com> References: <5893.1059721578@critter.freebsd.dk> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="bwzj1gAJAu3Kv4S1" Content-Disposition: inline In-Reply-To: <5893.1059721578@critter.freebsd.dk> User-Agent: Mutt/1.4i Organization: The FreeBSD Project Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-418-838-708 WWW-Home-Page: http://www.FreeBSD.org/ X-PGP-Fingerprint: 9A1B 8202 BCCE B846 F92F 09AC 22E6 F290 507A 4223 cc: arch@freebsd.org cc: Robert Watson Subject: Re: headsup: swap_pager.c X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 05 Aug 2003 00:55:30 -0000 --bwzj1gAJAu3Kv4S1 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Friday, 1 August 2003 at 9:06:18 +0200, Poul-Henning Kamp wrote: > In message , Robe > rt Watson writes: >> >> On Thu, 31 Jul 2003, Poul-Henning Kamp wrote: >> >>> I am in the process of changing the on-disk layout policy used by the >>> swap_pager. >> >> Do you anticipate any performance changes as a result of this change? > > Anticipate ? No. > Be surprised by improvements ? No. > > The striping code limits the I/O size for paging activity to a small > size. By laying out each disk sequentially in the "swap device" we > can increase that size to what our I/O system can lift. I'd expect a significant improvement. In our environment, throughput is almost linearly related to request size, so anything to increase the transfer size should improve the situation. Greg -- See complete headers for address and phone numbers --bwzj1gAJAu3Kv4S1 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.0 (FreeBSD) iD8DBQE/LwB8IubykFB6QiMRAu7zAKCQUYSG3oykG3OSzxgqNCd0MJwMRQCgl3zF c8S+h2Dw2BAcDbGz0XIj/Eg= =l0nw -----END PGP SIGNATURE----- --bwzj1gAJAu3Kv4S1--