SV: measuring 2.6.1 improvements

Tevfik Karagulle tevfik at
Tue Apr 27 22:49:55 GMT 2004


Results of a simple performance test on cygwin :

     333 files, 30 MB, flat directory, 3 test runs after ramp-up,

     Client : cygwin 1.5.9/ssh 3.8p on XP
     Server : cygwin 1.5.9/ssh 3.8p on windows 2000 Server (VMWARE)

     2.6.0 client to 2.6.0 server : 1 017 KB / sec
     2.6.1 client to 2.6.1 server : 1 206 KB / sec

     Performance improvement : 18,5 %

Very soon, I will update cwrsync ( with rsync 2.6.1
and ssh 3.8.1p1.

Rgrds Tev

-----Opprinnelig melding-----
Fra: at
[ at]Pa vegne av Wayne
Sendt: 27. april 2004 23:50
Til: Eric Whiting
Kopi: rsync at
Emne: Re: measuring 2.6.1 improvements

On Tue, Apr 27, 2004 at 02:14:10PM -0600, Eric Whiting wrote:
> I was watching the memory usage with a bash command line while loop
> while(true);do cat /proc/$pid/status |grep VmSize ;sleep 60;done

A very nice way to do that.  You might want to watch the VmRSS value,
which will tell you how much of the memory is resident.  A lower RSS
value means it will play nicer with other processes when memory is

> Wayne Davison wrote:
> > If you want to see an even larger improvement, compare the use of
> > the -H option (--hard-link) between versions (and note how much less
> > data is sent over the wire too).
> I'm confused. I don't think I have hard links in this tree. This
> improvement will not help me without hard links will it?

No, it won't help your situation.  If your goal is to fine-tune your
setup, ignore this comment.  If your goal is to measure improvements
in how rsync 2.6.1 handles memory compared to earlier versions, the
improvement when -H is used is very pronounced and worth a look-see.

> Not sure about my fork() -- here is the sys info:
> linux kernel 2.4.21

Linux has copy-on-write memory sharing for fork().  I'm not sure how to
ask /proc for the shared-memory value, though.  I usually use "top" for

To unsubscribe or change options:
Before posting, read:

More information about the rsync mailing list