On Jan 18, 2009, at 23:16 , Stefano Salsano wrote:
Cullen Jennings wrote:
>
> I'd like it if the requirements state that the solution be able to
> support a "make before break" style of handover even if it possible
> to use the solution to also do "break before make".
>
> Cullen <in my individual contributor role>
>
Dear Cullen (and all interested people),
we are revising the draft-niccolini-sipping-siphandover to take into
account the different comments and I am now tackling your comment.
We can simply take the comment literally, adding it in section 4:
The solution should support a "make before break" even if it is
possible
to use the solution to also do "break before make"
But if we dig a little bit into this problem... we can ask what is
meant
by "support": is it just "allow" or is it "exploit" ?
My wording was pretty vague. I think the requirements should address
if the solution needs the make before break as mandatory to implement
and optional to use or if it is optional to use. I don't care too much
one way or the other.
Therefore a second option is to include a mandatory requirement and an
optional one:
Mandatory req:
The solution should *allow* a "make before break" even if it is
possible
to use the solution to also do "break before make"
sure allow seems better
Optional req:
Using "make before break" should minimize packet loss. The solution
should also try to take care of the packet reordering due to the
vertical handover.
I agree we need text that gets to what does make before break even
mean from a requirements point of view so I like where you are going
here.
In my ideal world, if the handover went from transport A, to transport
B, the end user perceived quality during the handover would be no
worse that whichever of A or B had the worst quality around the time
of the handover. It's not easy to translate that to crisp
requirements. I'm bascially looking for a solution where the media on
B is being received by the receiver before the sender stops sending on
A.
In this respect we had some discussion among contributors to the
draft,
salvatore loreto wrote:
> how minimize the packet loss (or avoid it at all) and the re
ordering
> problem during a Vertical Handover! The question is there any
> protocol solution that can avoid it, or it is an hacking question?
I'd like to have some discussion and comments on this point: do we
take
care of requirements that relate to how the packets are handled in
order
to minimize loss and packet reordering ?
Cheers,
Stefano
Cullen <in my individual contributor role>
--
*******************************************************************
Stefano Salsano
Dipartimento Ingegneria Elettronica
Universita' di Roma "Tor Vergata"
Via del Politecnico, 1 - 00133 Roma - ITALY
http://netgroup.uniroma2.it/Stefano_Salsano/
E-mail : stefano.salsano@xxxxxxxxxxx
Cell. : +39 320 4307310
Office : (Tel.) +39 06 72597770 (Fax.) +39 06 72597435
*******************************************************************
_______________________________________________
Sipping mailing list https://www.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors@xxxxxxxxxxxxxxx for questions on current sip
Use sip@xxxxxxxx for new developments of core SIP
