Hi Greg, thanks for your comments, please see inline some initial considerations... Greg Bernstein wrote:
Hi Giovanni, this looks very interesting.
thx!
I noticed we both hit some similar issues ( see our draft http://www.ietf.org/internet-drafts/draft-bernstein-ccamp-wson-signaling-00.txt ), i.e., we both think there is a need for modulation type and FEC information. Since, in our draft we weren't looking at impairments we were using these more for end system compatibility and regenerator/wavelength converter compatibility. We looked at these more as "optical traffic parameters" so placed them in a Tspec object and used the attributes stuff for other information associated with wavelength assignment.Correct and the tspec was a solution we evaluate as well. In this case we preferred a newly defined (sub)TLV to have the possibility to signal more than one tuple (fec, modulation).
I noticed you expressed a concern on the amount of space used per lambda. Isn't it true that for wavelengths close to each other that the impairment (CD, PMD) parameters should be roughly the same so that we could group these with a "wavelength set"? We looked at a bit map encoding of a label set in both http://www.ietf.org/internet-drafts/draft-bernstein-ccamp-wson-signaling-00.txt and http://www.ietf.org/internet-drafts/draft-bernstein-ccamp-wson-info-00.txt such a mechanism could be helpful to reduce the space needed in the messages.The short answer is yes but probably not true in general. Using "wavelength set"(s) can leads to some approximation in evaluating your optical impairments while having parameters for each lambda is the general case. Another consideration, I'm wondering if even using "wavelength set" you might risk fragmentation in case your spectrum require too many sets.
Regards, G
Regards Greg B. Giovanni Martinelli (giomarti) wrote:Hi all, we submitted a new draft proposing lightpath setup through signaling extensions that take into account optical impairments. This allow asses optical feasibility in the context of wavelength switched optical networks. http://www.ietf.org/internet-drafts/draft-martinelli-ccamp-optical-imp-s ignaling-00.txt comments/suggestion are highly appreciated! Regards, Giovanni