From owner-freebsd-hardware@FreeBSD.ORG Fri Mar 26 18:54:38 2010 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 9C0F0106574B for ; Fri, 26 Mar 2010 18:54:38 +0000 (UTC) (envelope-from amikkili@ecn.purdue.edu) Received: from mx01.ecn.purdue.edu (smtp.ecn.purdue.edu [128.46.154.9]) by mx1.freebsd.org (Postfix) with ESMTP id B052B8FC16 for ; Fri, 26 Mar 2010 18:54:34 +0000 (UTC) Received: from delp-69-4.dhcp.ecn.purdue.edu (delp-69-4.dhcp.ecn.purdue.edu [128.46.69.231]) (authenticated bits=0) by smtp.ecn.purdue.edu (8.14.4/8.14.4) with ESMTP id o2QIc5ei019254 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT) for ; Fri, 26 Mar 2010 14:38:05 -0400 From: "Aravind K. Mikkilineni" To: freebsd-hardware@freebsd.org Date: Fri, 26 Mar 2010 14:38:04 -0400 User-Agent: KMail/1.9.9 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <201003261438.04897.amikkili@ecn.purdue.edu> X-ECN-MailServer-VirusScanned: by amavisd-new X-ECN-MailServer-Origination: delp-69-4.dhcp.ecn.purdue.edu [128.46.69.231] X-ECN-MailServer-SpamScanAdvice: DoNotScan Subject: Re: intellipark leads to high load cycle count X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 26 Mar 2010 18:54:38 -0000 On Tue, 23 Mar 2010 Michiel Boland wrote >Hi. I have one of those new green WD HDs with Intellipark, that is, they >like to park their heads every 8 seconds. As a result the load cycle count >grows at an alarming rate. (If I understand correctly this number should not >exceed about 300k or so.) > >Is there any way to turn off or increase the 8-second timeout? Or do I have >to bin the disk and use something better? I have some older GP drives which do not seem to ever park their heads. The newer GP drives I have behave as you described. I have had success either turning off the head parking or greatly reducing it by doing 'hdparm -S 0 -B 254 '. Either '-B 254' or -B 255' depending on the drive. I use these drives in a home file server, though, so I/O to/from the drives is either non-existant most of the time (heads stay parked) or there is continuous I/O (rdiff backups, etc.; heads stay unparked). -akm