From owner-freebsd-questions Thu Apr 22 15:53: 5 1999 Delivered-To: freebsd-questions@freebsd.org Received: from misha.cisco.com (misha.cisco.com [171.69.206.50]) by hub.freebsd.org (Postfix) with ESMTP id 0A20215A1A for ; Thu, 22 Apr 1999 15:52:44 -0700 (PDT) (envelope-from mi@misha.cisco.com) Received: (from mi@localhost) by misha.cisco.com (8.9.2/8.9.1) id SAA96045; Thu, 22 Apr 1999 18:49:36 -0400 (EDT) (envelope-from mi) From: Mikhail Teterin Message-Id: <199904222249.SAA96045@misha.cisco.com> Subject: Re: writing slower with SoftUpdates In-Reply-To: <19990422174301.B51036@dan.emsphone.com> from Dan Nelson at "Apr 22, 1999 05:43:02 pm" To: dnelson@emsphone.com (Dan Nelson) Date: Thu, 22 Apr 1999 18:49:36 -0400 (EDT) Cc: gjohnson@nola.srrc.usda.gov, questions@FreeBSD.ORG Reply-To: mi@aldan.algebra.com X-Mailer: ELM [version 2.4ME+ PL52 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Dan Nelson once wrote: > In the last episode (Apr 22), Glenn Johnson said: > > Interesting! I ran your script and found all three of my SCSI drives > > had write caching enabled. Is it the same situation with IDE drives > > and softupdates with regards to performance degradation? > Most IDE drives don't even support command queueing and have such > small internal buffers compared to equivalent SCSI drives (i.e. IBM's > Deskstar 16gb IDE has a 512k cache, while the Ultrastar 18gb SCSI has > a 2MB cache), so you're not going to see much of a decrease because of > the write cache. It is surprising, nothing in FreeBSD tries to take advantage of this huge caches... My first computer only had 128Kb of RAM in it... -mi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message