[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Re[2]: Performance of SVN on Windows and Linux

From: B Smith-Mannschott <bsmith.occs_at_gmail.com>
Date: Thu, 2 Apr 2009 19:56:20 +0200

On Thu, Apr 2, 2009 at 17:33, Andrey Repin <anrdaemon_at_freemail.ru> wrote:
> Greetings, David Weintraub!
> DW> The biggest  problem with Windows is the virus scanning software. When
> DW> you checkout 1000 files, Windows scans each of those for viruses. Then
> DW> of course, there are 1000 copies of all the files in the .svn
> DW> directory which Windows also scans. Then, there are all the properties
> DW> and other files in the .svn directories that also get scanned.
> It seems your imagination of Virus scanning schemes a bit outdated... Like, 8
> years old.
> Normal monitoring antivirus scan each file only once - at the creation or
> modification. And that does not take long time for single file.
> Then it forget about it. Reading from file, executing it will
> never trigger a scan unless you specifically told to do so, or other, non
> FS-related issues arise in process (Memory access to the different process,
> IRQ table manipulation, etc).

The topic of subversion client performance on Windows versus Linux was
discussed (to death) not long ago. Please see thread:


As to the effect of a (very badly configured) virus scanner as a
factor, please note something I contributed at that time:


The brain-dead McAffee configuration responsible for the performance
gap described in the above message is not something under the user's
control. It is imposed from without by the employer out of fear (of

// ben


To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-04-02 19:57:31 CEST

This is an archived mail posted to the Subversion Users mailing list.