From owner-freebsd-questions@FreeBSD.ORG Mon Apr 27 16:42:26 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6CBF51065672 for ; Mon, 27 Apr 2009 16:42:26 +0000 (UTC) (envelope-from korvus@comcast.net) Received: from mx04.pub.collaborativefusion.com (mx04.pub.collaborativefusion.com [206.210.72.84]) by mx1.freebsd.org (Postfix) with ESMTP id 389988FC15 for ; Mon, 27 Apr 2009 16:42:25 +0000 (UTC) (envelope-from korvus@comcast.net) Received: from [192.168.2.164] ([206.210.89.202]) by mx04.pub.collaborativefusion.com (StrongMail Enterprise 4.1.1.4(4.1.1.4-47689)); Mon, 27 Apr 2009 12:26:30 -0400 X-VirtualServerGroup: Default X-MailingID: 00000::00000::00000::00000::::1 X-SMHeaderMap: mid="X-MailingID" X-Destination-ID: freebsd-questions@freebsd.org X-SMFBL: ZnJlZWJzZC1xdWVzdGlvbnNAZnJlZWJzZC5vcmc= Message-ID: <49F5DCEF.4010208@comcast.net> Date: Mon, 27 Apr 2009 12:27:27 -0400 From: Steve Polyack User-Agent: Thunderbird 2.0.0.21 (X11/20090327) MIME-Version: 1.0 To: Ghirai References: <20090427191824.25e415e4.ghirai@ghirai.com> In-Reply-To: <20090427191824.25e415e4.ghirai@ghirai.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-questions@freebsd.org Subject: Re: quick vfs tuning X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 27 Apr 2009 16:42:26 -0000 Ghirai wrote: > The author suggested that increasing the default value of 8 to 16 > resulted in increased read speed, and that increasing it further > resulted in no noticeable performance gain. > > Personally, I've seen changes in vfs.read_max to provide anywhere from a 50-100% improvement in disk read performance. As you have also found, the "sweet spot" is sometimes higher than 16. I've seen significant benefits up to vfs.read_max=128 on *some* disk controllers. It really seems dependent on the disks/controllers, but setting 16 has always shown a significant improvement over the default 8 for me.