The Full Service Area Network model is no spring chicken. It's been on the books now for about four or five years. Various implementations have borrowed what they needed from FSAN, but it is seldom implemented, to the best of my knowledge, exactly the way it is spelled out by the consortium. Each vendor's adaptation is done (sometimes a lot) differently. What aspect are you referring to?
When FSAN was first conceived, wavelength division multiplexing was not widespread, limited to several lambdas, not dense, and these weren't even incorporated in the design at that time.
SONET was perceived to be the transport vehicle for everthing, and its scalability became a key gating factor for the number of homes and reach of any given ring topography.
Today, we might want to look at alternate approaches, namely using IEEE (10G) Gigabit Ethernet protocols in support of IP for certain types of applications, SONET for others, possibly native analog video for others, and native IP over lambda for others, still. WDM makes this possible over the same strand(s).
Another thought that has been crossing my mind lately about residential builds (which includes FSANs), is this. Multi-service networks that were conceived by the telco regime to combat the MSOs were ONLY designed to target residences, MDUs, apartments, maybe some small businesses, and the obligatory municipal building.
They were never, however, conceived as full service platforms to "all" potential end users along a route, only residential ones.
They excluded commercial enterprises and large works of various other persuasions as platforms for supporting all of "their" telecomm needs, as well. Again, a key design parameter was a one-for-one Plus match with the incumbent cable cos, and not to satisfy all telecomms requirements for all parties along a given route.
The reason for this stems from the mind set that was aimed at beating the cable cos, and a perceived limited number of applications and bandwidth allocations that would be needed to meet that end for every user, Or, every "passed home." You can call this the Home Dole, or the Residential Handout. It derives from a minimalist approach towards bandwidth, in an age when some highly paid folks worry about bandwidth glut.
And that's how the bandwidth allocation schemes and capacity budgets were meated out for FSAN, as well. Albeit, with a good deal more generous doling when it came to data services, vis a vis previous ISDN and V.34 alternatives, such as what we now call DSL Internet Access, which was only one of those SWAGs, when FSANs were first conceived, and still needed to prove itself. Well, I think it's proved itself.
In this age of soon-to-be multi-hundred wavelength wdm and robust routing and switching engines that would have been laughed at five years ago, however, I have to stop and wonder if the residential FSAN, as I've characterized it above, was the wisest decision for the incumbents (even the MSOs if they so chose to partake - as I sense T may be mulling over right now) to take.
Yes, I know... Monday morning quarterbacking, and so on, but it says something about the future potentials that lie in new builds, and these capabilities are something I've not seen or heard articulated anywhere thus far. New distribution networks don't have to be limited to Monday Night Football (and I don't care if it is in HDTV format) and MP3 downloads, IOW.
If what I've suggested here seems just a bit hairbrained, then consider that video is going IP, voice someday will too, and certainly Internet access will remain IP. IP, meaning to the end point, even if it does ride on top of ATM or Ethernet along the way.
Where vast differences in design existed in the past between analog coaxial systems and those which support just about everything else are beginning to wain, perhaps it's time to recognize that the divide is about to narrow to extinction (some day), and it's not too soon to begin thinking in terms of a unified approach in terms of both protocol and reach. Just some observations as I attempt to type myself to sleep.... |
