From owner-freebsd-fs@FreeBSD.ORG Thu Apr 11 17:08:56 2013 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id 5D81EFC for ; Thu, 11 Apr 2013 17:08:56 +0000 (UTC) (envelope-from nowakpl@platinum.linux.pl) Received: from platinum.linux.pl (platinum.edu.pl [81.161.192.4]) by mx1.freebsd.org (Postfix) with ESMTP id 239A4AE3 for ; Thu, 11 Apr 2013 17:08:55 +0000 (UTC) Received: by platinum.linux.pl (Postfix, from userid 87) id CAB4647E29; Thu, 11 Apr 2013 19:08:54 +0200 (CEST) X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on platinum.linux.pl X-Spam-Level: X-Spam-Status: No, score=-1.3 required=3.0 tests=ALL_TRUSTED,AWL autolearn=disabled version=3.3.2 Received: from [10.255.1.2] (unknown [83.151.38.73]) by platinum.linux.pl (Postfix) with ESMTPA id 9DEB047E24 for ; Thu, 11 Apr 2013 19:08:54 +0200 (CEST) Message-ID: <5166EE14.1090806@platinum.linux.pl> Date: Thu, 11 Apr 2013 19:08:36 +0200 From: Adam Nowacki User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/20130328 Thunderbird/17.0.5 MIME-Version: 1.0 To: freebsd-fs@freebsd.org Subject: Re: ZFS slow reads for unallocated blocks References: <5166EA43.7050700@platinum.linux.pl> <5166ECBA.1090005@gmail.com> In-Reply-To: <5166ECBA.1090005@gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 11 Apr 2013 17:08:56 -0000 On 2013-04-11 19:02, Chuck Burns wrote: > On 4/11/2013 11:52 AM, Adam Nowacki wrote: >> This one is quite weird - reads from files that were created and resized >> with ftruncate (so no actual data on disk) are considerably slower and >> use more CPU time than files with data. If compression is enabled this >> will also affect files with long runs of zeroes as ZFS won't write any >> data to disk in this case. There is no I/O on the pool during the read >> tests - all fits into 10GB ARC. >> >> FreeBSD storage 9.1-RELEASE FreeBSD 9.1-RELEASE #0: Sat Feb 23 15:51:26 >> UTC 2013 root@storage:/usr/obj/usr/src/sys/GENERIC amd64 > > > Sounds like it could be a CPU bottleneck? How about some cpu info? > > hw.model: AMD FX(tm)-4100 Quad-Core Processor hw.ncpu: 4 But this doesn't matter much as CPU time should be compared relative to other tests - 1 second versus 15 seconds on the same CPU.