[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Netconf on Debian - benchmarking results



Hi RS,

I would definitely be interested in seeing your benchmark results, when 
you can publish them.

Cheers,
BobN

---- Original message ----
>Date: Mon,  9 May 2005 23:56:44 +0200
>From: Radu.State@loria.fr  
>Subject: Re: Netconf on Debian  
>To: j.schoenwaelder@iu-bremen.de
>Cc: Vincent Cridlig <Vincent.Cridlig@loria.fr>, netconf@ops.ietf.org, 
Radu State <Radu.State@loria.fr>
>
>
>Thanks for the comments.
>
>The current release has an optional encryption and compression module, 
which is
>mere optional. Its just an implementation of our IEEE/IFIP Integrated
>Management IM  2005 paper. Anyway, its usage is optional and 
transparent,
>basically if an application does not want to use them, everyhing 
should work
>just fine.
>
>We will support only the SSH binding in the short future: for the 
moment one has
>to rely on standart TCP/SSH tunneling, but we will add a programmatic 
support
>shortly.
>
>Regarding the frame marking, your are right. We will change it 
tomorrow, we
>forgot to change it. Thanks for pointing it out to us.
>
>There are some additional (with respect to the draft) features in the 
agent
>(like role based access control, XSLT pushing/rendering, etc), but 
they are
>optional in usage and transparent for a normal usage.
>
>Thanks one more time and please send us more comments/bugs 
notification.
>
>Rearding other potential interesting experience for the WG, we can 
post (if
>there is interest for, after IM 2005) a technical report on some 
implemented
>extensions  to the draft and benchmarking results which are rather 
surprising.
>
>RS
>
>
>
>Selon Juergen Schoenwaelder <j.schoenwaelder@iu-bremen.de>:
>
>> On Mon, May 09, 2005 at 04:14:07PM +0200, Vincent Cridlig wrote:
>>
>> > The Madynes research team has just released a new implementation 
of a
>> > Netconf agent (YencaP).
>>
>> [...]
>>
>> > Any questions, suggestions for improvement or comments are welcome.
>>
>> Like the previous yenca release (which was written in C), you do not 
seem
>> to support any of the three transport mappings that are discussed in 
the
>> working group. The first yenca release used a TLS stream (but did not
>> get the framing quite right) while this release runs netconf over TCP
>> (but with a frame marking mechanism that is different from the one 
used
>> in the netconf over ssh specification). In addition, you seems to
>> support encryption and compression above the netconf rpc layer, which
>> does not match the approach taken by the WG (which is to leave 
encryption
>> and compression to the transport).
>>
>> I did not look through the other parts of the code but I am wondering
>> how much netconf there actually is in yenca. I also like to know why 
you
>> did depart from the netconf transport specifications (and perhaps 
other
>> parts of netconf) and whether there is anything that the WG can learn
>> from this.
>>
>> /js
>>
>> --
>> Juergen Schoenwaelder		    International University 
Bremen
>> <http://www.eecs.iu-bremen.de/>	    P.O. Box 750 561, 28725
Bremen, Germany
>>
>
>
>
>
>
>
>--
>to unsubscribe send a message to netconf-request@ops.ietf.org with
>the word 'unsubscribe' in a single line as the message text body.
>archive: <http://ops.ietf.org/lists/netconf/>

--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>