From owner-freebsd-current@FreeBSD.ORG Sat Dec 11 14:24:17 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2DABC16A4CE; Sat, 11 Dec 2004 14:24:17 +0000 (GMT) Received: from web.portaone.com (web.portaone.com [195.70.151.35]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6677643D5A; Sat, 11 Dec 2004 14:24:16 +0000 (GMT) (envelope-from sobomax@portaone.com) Received: from [192.168.1.26] ([192.168.1.26]) (authenticated bits=0) by web.portaone.com (8.12.11/8.12.11) with ESMTP id iBBEOD0a093231 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sat, 11 Dec 2004 15:24:14 +0100 (CET) (envelope-from sobomax@portaone.com) Message-ID: <41BB0305.7000206@portaone.com> Date: Sat, 11 Dec 2004 16:24:05 +0200 From: Maxim Sobolev Organization: Porta Software Ltd User-Agent: Mozilla Thunderbird 1.0 (Windows/20041206) X-Accept-Language: en-us, en MIME-Version: 1.0 To: phk@freebsd.org Content-Type: text/plain; charset=KOI8-U; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: ClamAV 0.80/589/Wed Nov 17 13:38:41 2004 clamav-milter version 0.80j on www.portaone.com X-Virus-Status: Clean cc: "current@freebsd.org" Subject: GBDE write performance really sucks X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Dec 2004 14:24:17 -0000 Hi Poul, I have noticed that GBDE write performance is very low, even on modest hardware (3.2 P4 with 1MB cache and 7200 IDE HDD): $ sudo dd if=/dev/zero of=/dev/ad0e.bde bs=128k 29491200 bytes transferred in 31.959903 secs (922756 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 229+0 records in 228+0 records out 29884416 bytes transferred in 32.376493 secs (923028 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 232+0 records in 231+0 records out 30277632 bytes transferred in 32.800038 secs (923097 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 238+0 records in 237+0 records out 31064064 bytes transferred in 33.637685 secs (923490 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 242+0 records in 241+0 records out 31588352 bytes transferred in 34.204824 secs (923506 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 247+0 records in 246+0 records out 32243712 bytes transferred in 34.898871 secs (923918 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 258+0 records in 257+0 records out 33685504 bytes transferred in 36.440052 secs (924409 bytes/sec) load: 0.00 cmd: dd 3289 [physwr] 0.00u 0.02s 0% 832k 262+0 records in 261+0 records out 34209792 bytes transferred in 36.990503 secs (924826 bytes/sec) gstat(8) shows that ad0e.bde is nearly 100% busy. I guess that exsessive seeks back and forth are knocking performance down, but are there any plans to address this issue? Since practical usefullness of the file system that can't even sustain 1MB/s on modest hardware is quiestionable. Regards, Maxim