From owner-freebsd-questions@FreeBSD.ORG Wed Mar 24 20:48:09 2010 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 099141065670; Wed, 24 Mar 2010 20:48:09 +0000 (UTC) (envelope-from bfriesen@simple.dallas.tx.us) Received: from blade.simplesystems.org (blade.simplesystems.org [65.66.246.74]) by mx1.freebsd.org (Postfix) with ESMTP id C24138FC1B; Wed, 24 Mar 2010 20:48:08 +0000 (UTC) Received: from freddy.simplesystems.org (freddy.simplesystems.org [65.66.246.65]) by blade.simplesystems.org (8.13.8+Sun/8.13.8) with ESMTP id o2OKlx3Y015304; Wed, 24 Mar 2010 15:47:59 -0500 (CDT) Date: Wed, 24 Mar 2010 15:47:59 -0500 (CDT) From: Bob Friesenhahn X-X-Sender: bfriesen@freddy.simplesystems.org To: Dan Nelson In-Reply-To: <20100324175546.GF12330@dan.emsphone.com> Message-ID: References: <20100324175546.GF12330@dan.emsphone.com> User-Agent: Alpine 2.01 (GSO 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.2 (blade.simplesystems.org [65.66.246.90]); Wed, 24 Mar 2010 15:47:59 -0500 (CDT) Cc: freebsd-fs@freebsd.org, Dan Naumov , freebsd-questions@freebsd.org Subject: Re: tuning vfs.zfs.vdev.max_pending and solving the issue of ZFS writes choking read IO 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: Wed, 24 Mar 2010 20:48:09 -0000 On Wed, 24 Mar 2010, Dan Nelson wrote: > > I had similar problems on a 32GB Solaris server at work. Note that with > compression enabled, the entire system pauses while it compresses the > outgoing block of data. It's just a fraction of a second, but long enough > for end-users to complain about bad performance in X sessions. I had to > throttle back to a 256MB write limit size to make the stuttering go away > completely. It didn't affect write throughput much at all. Apparently this was a kernel thread priority problem in Solaris. It is apparently fixed in recent versions of OpenSolaris. The fix required adding a scheduling class which allowed the kernel thread doing the compression to be less than the priority of normal user processes (such as the X11 server). Bob -- Bob Friesenhahn bfriesen@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/