From owner-freebsd-fs@freebsd.org Wed Aug 19 02:28:53 2015 Return-Path: Delivered-To: freebsd-fs@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 1D4D39BD432 for ; Wed, 19 Aug 2015 02:28:53 +0000 (UTC) (envelope-from bfriesen@simple.dallas.tx.us) Received: from blade.simplesystems.org (blade.simplesystems.org [65.66.246.74]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id BD017FD6 for ; Wed, 19 Aug 2015 02:28:52 +0000 (UTC) (envelope-from bfriesen@simple.dallas.tx.us) Received: from freddy.simplesystems.org (freddy.simplesystems.org [65.66.246.65]) by blade.simplesystems.org (8.14.4+Sun/8.14.4) with ESMTP id t7J2GAQV009049; Tue, 18 Aug 2015 21:16:10 -0500 (CDT) Date: Tue, 18 Aug 2015 21:16:10 -0500 (CDT) From: Bob Friesenhahn X-X-Sender: bfriesen@freddy.simplesystems.org To: javocado cc: FreeBSD Filesystems Subject: Re: Optimizing performance with SLOG/L2ARC In-Reply-To: Message-ID: References: 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]); Tue, 18 Aug 2015 21:16:11 -0500 (CDT) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 Aug 2015 02:28:53 -0000 On Tue, 18 Aug 2015, javocado wrote: > I've been trying to optimize and enhance my ZFS filesystem performance > (running FreeBSD 8.3amd) which has been sluggish at times. Thus far I have > added RAM (256GB) and I've added an SLOG (SSD mirror). The RAM seems to > have helped a bit, but not sure if the SLOG was of much help. My vdev is > decently busy, with writes and reads averaging at 100 per second with > spikes as high as 500. Lots of interesting questions. You have not told us the use case for your system, or the zpool layout. Your comment about 'vdev is decently busy' causes me to think that perhaps you have just one and that more vdevs will (at least) linearly improve over-all performance. > 1. based on the output above, I believe a larger ARC may not necessarily > benefit me at this point. True? It looks like your ARC is doing well. > 2. Is more (L2)ARC always better? No. If (L2)ARC ends up empty, then it is wasted. > 3. I know it's a good idea to mirror the SLOG (and I have). Do I understand > correctly that I do not need to mirror the L2ARC since it's just a read > cache, nothing to lose if the SSD goes down? That is my understanding. Everything in the (L2)ARC is also on the pool disks. Look into re-architecting your pool. It is not clear what type of reads/writes are taking place but if these are random access to pool disks and you are using a raidzN organization, then you may be bottlenecked on disk write I/Os. This is the important thing to determine. If you were bottlenecked on async disk write I/Os and your slog is relatively idle, then you may benefit most from more vdevs and possibly more disks. Decreasing read I/Os will help, but can only go so far. Use mirrors if you can afford it. Bob -- Bob Friesenhahn bfriesen@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/