|
Hi Greg, I’ll have a look to the
ID thanks. Best regards Diego Diego Caviglia From: Greg Bernstein [mailto:gregb@grotto-networking.com] Hi Diego, looking at MPLS and GMPLS particularly
applied to SONET/SDH/G.709 we see that most switches are assumed to be able to convert
any ingress label to any egress label (within say some range). In the
case of lambda switching without wavelength converters (your O-E-O transponder)
we do not have this capability (labels map to lambdas). Hence this could also
be interpreted as an additional constraint on the switch. I don't think we
currently have a way to represent this in our routing protocols. Please
take a look at the draft http://www.ietf.org/internet-drafts/draft-bernstein-ccamp-wavelength-switched-00.txt
that Young Lee and I put together. It seems to us that some extensions to
GMPLS maybe necessary to address your application. If I interpret it
correctly ;-)
Hmmmmmm not sure my Understanding of the lambda switching is what I’ve
called spatial switching that is lambda1 portA à lambda1
portB what is not clear to me is how can be advertised an OEO transponder that
can only perform frequency switching lambda1 à lambda 2. BR Diego -----Original Message----- Hi Diego. I believe we should refer to the Holly RFC 3945,
chapter 1, verse 2: - "Lambda Switch Capable" interfaces
"can operate at the level of an *individual wavelength*" [or a
"group of wavelengths"], meaning that you manipulate values of
wavelengths (as AU-4 numbers [or AU-4 ranges] from an SDH portA to SDH portB),
like in a ROADM; - "Fiber-Switch Capable" interfaces
"can operate at the level of a single or multiple *fibers*", meaning
*spatial switching* where you don't consider the type of signal that ports
convey (could be anything like a black and white signal, a wavelength, a WDM
multiplex, some optical packets...), like in a OOO PXC. To stick with strict terminolgy: lambda = wavelength
= (speedOfLight / frequency) So if you need to do "frequency
switching", then it is the so called "lambda switching". :-) Anyway, this is my understanding, so if I'm wrong or
if it's a vocabulary issue because you find that terms are inappropriate, then
we'd better ask father Adrian and sister Deborah. Cheers, Julien -----Original Message----- From: Diego Caviglia (GA/ERI) [mailto:diego.caviglia@ericsson.com]
Hi Julien,
Actually not the PXC I had in mind is able to switch a single lambda I didn't
but the mux/demux In the picture sorry. The point I failed to illustrate is the ambiguity of
the term "Lambda Switch Capable" given that there two possible ways
to switch a lambda. The first one is the spatial one: (Lambda1 portA) -->
(Lambda1 portB) this is the way an all optical switch works and this why there
is the lambda continuity constraint in photonic networks. The second one is the frequency switching: (Lambda1
portA) --> (Lambda2 portA) this switching can be done via a transponder
(OEO) device. Of course is possible to mix the two switching
having (Lambda1 portA) --> (Lambda2 portB) My impression is that the definition "Lambda
Switch Capable" refers to the spatial switching and thus I don't know how
to model the fact that after/before a photonic matrix I have a
transponder. I hope I've made my question clearer. Best Regards Diego -----Original Message----- From: MEURIC Julien RD-CORE-LAN [mailto:julien.meuric@orange-ftgroup.com]
Sent: martedì 3 luglio 2007 19.21 To: Diego Caviglia (GA/ERI); ccamp@ops.ietf.org Subject: RE: Switching Capability of Photonic Links
with Transponder Hi Diego. If I understand correctly, your "lambda
switch" by itself is a PXC that has only "Fiber-Switch Capable"
interfaces. Then, you add lambda-conversion cards to it. So, correct me if I'm
wrong (you or anyone else), but whether you do a lambda conversion
inside a card or in a core matrix, this new interface on your global
device is able to work on lambdas anyway [(lambda 1, port A) -->
(lambda2, port B)]. As a result, you need to advertise your most flexible
capability, which is "Lambda Switch Capable". If you used "FSC", you wouldn't be able to
control your "lambda swapping" card, as LSPs are like lists of
fibers and labels aren't wavelengths but ports. But maybe I didn't get your actual issue. My 2 cents, Julien ________________________________ From: owner-ccamp@ops.ietf.org
[mailto:owner-ccamp@ops.ietf.org]
On Behalf Of Diego Caviglia (GA/ERI) Hi all,
I've a doubt about how to model the following situation. +-----------------+
|
|-------+
|
| OEO |
| Lambda |-------+
| Switch |
|
|
|
| +-----------------+ The node itself is able to cross connect only the
Lambda while the interface has a OEO transponder that is able to
change the lambda frequency. In this case there are two
different 'switching capability' the spatial one that is performed by the switch
(lambda 1, port A) --> (lambda1, port B) and the frequency switching is
done by the OEO transponder. Witch kind of interface switching
capability I have to advertise? BR Diego Diego Caviglia Product PA Broadband BNET Marconi S.p.A Ericsson Global Via Anagnina,203 0018, Roma , www.ericsson.com
<http://www.ericsson.com/>
Office: +39 010 600 3736 Fax: +39 010 600 3493 Email: diego.caviglia@ericsson.com
This communication is confidential and intended
solely for the addressee(s). Any unauthorized review, use,
disclosure or distribution is prohibited. If you believe this message has been
sent to you in error, please notify the sender by replying to this
transmission and delete the message without disclosing it. Thank you. E-mail including attachments is susceptible to data
corruption, interception, unauthorized amendment, tampering and
viruses, and we only send and receive emails on the basis that we are not
liable for any such corruption, interception, amendment, tampering or
viruses or any consequences thereof.
-- ===================================================Dr Greg Bernstein, Grotto Networking (510) 573-2237 |