Re: Congestion in M3UA/SUA and the use of the SCON

"Brian F. G. Bidulock" <[email protected]>
Newsgroups gmane.ietf.sigtran
Organization http://www.openss7.org/
Message-ID <[email protected]>
Haresign,

Haresign Lincoln wrote:                                                            (Tue, 11 Oct 2005 14:01:58)
> Brian,
> 
> I'm not sure what your point is. Detection of congestion is generally
> considered to be "implementation dependent".  One of the functions of
> SubSystem Management is to manage the state of the local subsystem.  One
> of these states is congested.  The detection of congestion towards a
> particular subsystem is implementation dependent.  I didn't say it was
> in the SCCP user.

But you keep saying that the ASP (SCCP-User) does not know its point code.
(Even though according to Q.711, an SCCP-SAP contains an MTP-SAP (point
code).)

Do you wish to place SCCP congestion detection at the SCCP user (yours
of which it appears doesn't not know its own point code) or with the
SCCP (that must not only know its point code, but must be able to
determine all other "concerned" point codes)?

By allowing an ASP to send SCON at all, we are allowing part of the SCCP
to be dragged across the line from SG to ASP.  This is already necessary
to some degree in the multiple SG as STP scenarios.  However, if the ASP
design wishes to drag some of the SCCP procedures accross into the ASP,
it must also be able to perform those SCCP procedures and is no longer a
strict SCCP-User.  Nevertheless, it is abuse of those procedures at the
ASP that makes it unwise to require the SG to do anything with SCON
received from an ASP.

BTW, the mechanism for detection of congestion within the SCCP
implementation is, of course, dependent upon the internal particulars of
that SCCP implementation.  However, "implementation specific" and
"implementation dependent" in the SCCP specifications does not affect
functional placement of the congestion functions (they are clearly
within the SCCP), nor does it allow one to avoid following the
congestion procedures (which include sending a message that includes
both point codes).

--brian

-- 
Brian F. G. Bidulock
[email protected]
http://www.openss7.org/
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.