Query on performance

[Juli Mallett]

Hi Madhan,

There are a few places you can pull together information on the
theoretical results and on real-world results; I'd suggest starting
with these:

http://wanproxy.org/performance.shtml
http://lists.wanproxy.org/pipermail/wanproxy-wanproxy.org/2014-October/001551.html
http://lists.wanproxy.org/pipermail/wanproxy-wanproxy.org/2011-March/001254.html
http://wanproxy.org/ssh.shtml

Note that none of those address the fact that WANProxy's encoding
between nodes has gotten more sophisticated with time, and now handles
things like retransmitting data that the remote peer has forgotten,
etc., so there is additional overhead worth thinking about in the
abstract.  And I've never updated the "performance" page to reflect
_any_ changes since the initial public version of the protocol.

Beyond that, and comparisons to proprietary WAN Optimization
solutions, or other deduplication solutions, I'd be hard-pressed to
find anything public.  I'd encourage doing the comparisons yourself,
if you're set up to do so!

Empirical tests are always more meaningful than the theoretical cases
anyway.  It's part of why the "tack" tool is included — so you can do
deduplication by hand with real data and see what it looks like in a
controlled setting.  "xcdump" isn't really useful anymore, sadly.
Really, we need a Wireshark dissector probably, and I haven't written
one in years, and am unlikely to make the time to as for my own
purposes I find the wire protocol easy enough to keep in my head.  I'm
a self-serving developer in that way.

Thanks,
Juli.

On Sat, Jan 30, 2016 at 10:23 PM, Madhan <madhan.mepco at gmail.com> wrote:
> Hi ,
> I am Madhan from Samsung research . I am studying more detail on the
> compression and data optimization .
> Do we have any reference of wan proxy performance ? Any white paper
> which talks about the percentage of compression and data saved with
> respect to the normal connection and/or other existing solution .
>
> --
> Regards,
> Madhan Raj K ,
> Senior Software Engineer - Android R&D ,
> Samsung Research Institute - Bangalore .
> _______________________________________________
> wanproxy mailing list
> wanproxy at lists.wanproxy.org
> http://lists.wanproxy.org/listinfo.cgi/wanproxy-wanproxy.org