X-Recipient: archive-cygwin AT delorie DOT com X-Spam-Check-By: sourceware.org Date: Thu, 8 May 2008 11:23:11 -0400 From: Christopher Faylor To: cygwin AT cygwin DOT com Subject: Re: Performance of 1.7 Message-ID: <20080508152311.GD32445@ednor.casa.cgf.cx> Reply-To: cygwin AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.16 (2007-06-09) Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com On Wed, May 07, 2008 at 10:56:19PM -0600, Rob Bosch wrote: >I've been testing cygwin 1.7 on Server 2008 with rsync. First, the install >went flawlessly. I'm just using a standard path of E:\cygwin...no fstab >issues. > >Compiling rsync worked without a hitch. The rsync version is 3.02 with the >preallocate patch applied. The patch takes advantage of the posix_fallocate >feature. > >There have been two major improvements I've seen with the 1.7 build on >Server 2008: > >- The long file name errors that used to always be a problem are no longer >there which is great news. >- The overall I/O performance is much higher. I don't have hard statistics >but have been watching throughput on the SAN connected to the server. With >old cygwin and Server 2003 I was lucky to get sustained throughput of >40M-50M. Now I'm seeing throughput of 100M-120M all the time with bursts >much higher. It is possible the performance gain is helped by Server 2008 >as I have not attempted to run the old cygwin on 2008 for a true comparison. >I think a lot of it is also in the cygwin version since I'm sending the >email to that list. :) > >Sorry I don't have any more hard data as we continue testing 1.7. Thought >I'd at least let you know some of the nice things I'm seeing. If you're using pipes in any capacity I would expect that you'd see some improvement in throughput since I rewrote the I/O handling there to remove the old head-standing which was necessary for Windows 9x. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/