& cplSiteName &

Verizon: WDM-PON Still Sounds Expensive

Craig Matsumoto
9/13/2010

LAS VEGAS -- Verizon Communications Inc. (NYSE: VZ) intends to issue a next-generation PON request for information (RFI) later this month, and while WDM-PON hasn't been ruled out, its expense might make it a long shot.

When Verizon starts making its choice, "one of the key features would have to be cost parity with existing technologies," said Shweta Jain, a technical lead in Verizon's wireline access technologies division.

Of course, Verizon uses GPON, and that makes XG-PON a favorite for its next fiber phase; in fact, Jain's bio describes her as the XG-PON prototype testing lead.

But she was speaking Sunday at the WDM-PON Forum Workshop, a day-long event piggybacked onto the 2010 FTTH Conference and Expo. So, the question of WDM-PON was fair game. (See WDM-PON Gets Its Day.)

Organizers were taking their first stab at this workshop, and they were pleased to see about 130 attendees turn up -- not bad for a Sunday morning. In Las Vegas. On the first weekend of football, AND practically next door to the World Championship of Fantasy Football.

Cost was one of the day's lingering topics, because WDM-PON has a reputation for being expensive. But Jain's comments, and a moderator's insistence on repeatedly comparing GPON to WDM-PON, got some hackles up.

WDM-PON will always look expensive next to GPON because WDM-PON is a point-to-point technology and GPON delivers shared bandwidth, said Wim te Niet, an LG-Ericsson Co. Ltd. vice president of sales. He didn't think the two should be compared -- although, as Verizon's example shows, that comparison is happening.

James Salter, founder and chief strategy officer of Atlantic Engineering Group -- a company that builds networks for small operators and municipalities -- said he's quoted only one WDM-PON job, with costs that came to $3,500 per subscriber. His GPON examples, by contrast, had figures more like $1,700 or $2,000 per subscriber.

Salter is a WDM-PON fan, though, and he thinks it's better justified by considering the cost per 1 Mbit/s of bandwidth. Because the WDM-PON job was designed for 1 Gbit/s services, that figure came out to around $3.50, compared with about $25 for a GPON buildout.

No one at the conference harbored illusions of a quick WDM-PON revolution. In one of the morning's first talks, analyst Jeff Heynan of Infonetics Research Inc. said any big ramp-up won't start until 2014, at which point the market size will be $88 million. That time frame went unchallenged for the rest of the day.

— Craig Matsumoto, West Coast Editor, Light Reading

(17)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Page 1 / 2   >   >>
Pete Baldwin
Pete Baldwin
12/5/2012 | 4:23:56 PM
re: Verizon: WDM-PON Still Sounds Expensive


That one WDM-PON quote that AEG's Salter mentioned is still active -- that is, the customer's still deciding.


There was also a WDM-PON buildout described by Mike Smeltzer of U.of Illinois, Champaign-Urbana. It would be 7 rings, each one going through 2 common points, reaching to many of the surrounding underserved areas.  And if they do decide to use WDM-PON, they'll probably be the only Recovery Act-funded project to do so.


So, the stuff is out there and getting attention in the U.S.


I do wonder, though ... Recovery Act funding meant there was free money to build one's dream optical network -- seems like a great time to propose WDM-PON, but I haven't heard of anyone who did, besides Champaign-Urbana.

Duh!
Duh!
12/5/2012 | 4:23:55 PM
re: Verizon: WDM-PON Still Sounds Expensive


Of all the speakers, did any of them articulate an actual value proposition?  You know, as in "here's why a WDM PON solution will have a shorter time to positive cash flow than GPON or GE-PON" or "here's a network application that WDM PON is uniquely suited to serve"?   Did anybody attempt to rebut my shared bandwidth argument from a few months ago? Did anybody talk about roadmaps to parity in OLT port density, or OLT port cost or remote node cost or ONT cost? 


 

paolo.franzoi
paolo.franzoi
12/5/2012 | 4:23:54 PM
re: Verizon: WDM-PON Still Sounds Expensive


 


There is an error in the statement that WDM-PON is a point to point technology.  That is not factually accurate.  There are SOME point to point WDM-PON technologies - by this it is meant that each endpoint has its own wavelength.  However, there are other WDM-PON technologies that are multi-wavelength where multiple endpoints are still share a wavelength.


The intent of the former is theoretically a slightly lower cost build than actual point to point fiber.  The intent of the latter is a massive restructuring of how PONs are constructed (imagine an OLT with 100K subs for example).


seven

paolo.franzoi
paolo.franzoi
12/5/2012 | 4:23:52 PM
re: Verizon: WDM-PON Still Sounds Expensive


Craig,


Did anyone compare WDM-PON to just point to point fiber connections in the mode they talked about?


That is why I never really understood the advantage of the point to point systems.  What you gain is:


Fewer Cables


What you lose is:


More complex optics


Low volume components


They certainly can not say that a wavelength selected point to point system is simpler than running a L2 GigE switch with lots of ports.


seven


 

Pete Baldwin
Pete Baldwin
12/5/2012 | 4:23:52 PM
re: Verizon: WDM-PON Still Sounds Expensive


OK, point taken, 7.  (No pun intended.)  No one really talked about the multiwavelength case on Sunday, fwiw.


> There is an error in the statement that WDM-PON is a point to point technology.  That is not factually accurate.  There are SOME point to point WDM-PON technologies - by this it is meant that each endpoint has its own wavelength.  However, there are other WDM-PON technologies that are multi-wavelength where multiple endpoints are still share a wavelength.


 

Pete Baldwin
Pete Baldwin
12/5/2012 | 4:23:52 PM
re: Verizon: WDM-PON Still Sounds Expensive


> Of all the speakers, did any of them articulate an actual value proposition?


I don't think you'd have heard anything that would change your mind, if that's what your asking.  :)


Primary arguments for WDM-PON remain the same, among them, lower costs due to eliminating a lot of outside-plant power, maintenance, and upgrading.  A couple of specific applications did come up -- some enterprises prefer to be on a network where they get dedicated wavelengths, for instance.


They did address the cost on the ONT side, particularly from the optical components point of view. AOI and Luxpert gave talks about injection-locked Fabry Perot lasers, for instance.

Pete Baldwin
Pete Baldwin
12/5/2012 | 4:23:38 PM
re: Verizon: WDM-PON Still Sounds Expensive


Seven -- No, there wasn't a lot of discussion about the specifics of other point-to-point options.


Lower-volume components is certainly one handicap they're aware of; that's part of the whole discussion about high costs.  More complex optics -- there seems to be a tradeoff of moving more of the optics (& possibly the complexity) to the edges -- although that brings up density questions, as mentioned here. 


What about an AWG instead of a splitter in the middle of the network -- the AWG seems simpler & is obviously passive, but has issues of its own (which I'm still meaning to discuss in a followup).


Certainly some good points to think about, in any event. Thanks.

Pete Baldwin
Pete Baldwin
12/5/2012 | 4:23:38 PM
re: Verizon: WDM-PON Still Sounds Expensive


A quick correction -- that's an RFI that Verizon is readying, not an RFP.  I'll fix it in the story shortly.

Duh!
Duh!
12/5/2012 | 4:23:37 PM
re: Verizon: WDM-PON Still Sounds Expensive


Seven,


The case for PON (either TDM or WDM) over point-to-point turns on the benefit of reduced fiber count in the feeder plant (from CO to remote node).  It is true that incremental construction cost for high count cable is small (relative to low count cable) in true greenfields deployments.   However,  are often cases where FTTH deployments can end up sharing an existing route, and when that is the case,  a smaller diameter cable can be the difference between sharing existing duct and trenching new duct.  Reduced fiber count also means fewer CO terminations, less congestion at CO fiber distribution frames, fewer opportunities for crafts errors in FDF operations.  Finally, restoral after a cable cut is much faster, due to fewer splices, more subs restored per splice, easier handling of smaller count cable.


The further case for TDM PON is that the full shared capacity of the feeder can be exploited due to statistical gain.  Also, higher OLT port density can be achieved.


Incidentally, there are FTTH schemes which involve active remote nodes.    The  problems, obviously are powering, maintenance and environmental hardening.


Also to Craig's point, athermal AWGs are significantly more complex than either fused fiber or PLC-based splitters.  If nothing else, all the schemes that I've seen for thermal compensation involve additional processing steps.  And since splitters are already so far down the learning curve, it would take tremendous volumes to get AWGs to approach price parity. 


 

paolo.franzoi
paolo.franzoi
12/5/2012 | 4:23:37 PM
re: Verizon: WDM-PON Still Sounds Expensive


 


Well, an AWG is how point to point colorless WDM-PONs exist.  They use the AWG to select the color to go to a home and then at least the old Nortel solution (aka ex-Novera) locks its transmitter to the received light frequency to provide a colorless ONT.


Here is the thing.  If you run point to point standard ethernet, you don't need any of that.  You just run more fibers in the bundle.  You have no AWGs and no splitters.  So, the point to point WDM-PONs seem to be complex way of reducing the cable bundle.  They will also claim that you need fewer optical ports, but that is a bit of a red herring.  You still need a laser per wavelength, so you only really save on external connectors and space for those.


A TDM-PON uses 1 laser to talk to multiple endpoints and that is where the savings are.  A Point to Point system has a 1 - 1 mapping so there is no laser savings.  I did some looking at wide-bandwidth, high power laser sources awhile back.  Seemed impractical at the time but the technology may have change.


So, what I am confused by as to the business case - which is where the point of the article was:


1 - Are the comparisons to TDM-PONs


2 - Are the comparisons to Point to Point Fiber


3 - Are the comparisons to Copper


So, all the fiber examples can have passive outside plants.  That is why I am unclear.


seven


PS -


ftp://ftp.cordis.europa.eu/......






 

Page 1 / 2   >   >>
Featured Video
Upcoming Live Events
November 5, 2019, London, England
November 7, 2019, London, UK
November 14, 2019, Maritim Hotel, Berlin
December 3-5, 2019, Vienna, Austria
December 3, 2019, New York, New York
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events