Conventional wisdom used to have it that to see the future of fixed telecom, you should look East to Japan or South Korea for Fiber/Optical Fiber access – or to India for, say, Mobile WiMax. Now that Verizon Communications Inc. (NYSE: VZ) is starting to roll out fiber access, even the U.S. is worth a look.
But to know when the future really has arrived, eagle-eyed sceptics should perhaps look at what BT Group plc (NYSE: BT; London: BTA) is doing in the U.K. This national incumbent has notoriously held out against blowing the shareholder dividend on fiber access, so when the company finally said it might roll out fiber in a big way (as it did in July 2008, although with a huge financial proviso), it signaled that fiber access is for real, even if not a reality for most Brits.
But there is a serious point here. New technology and infrastructure costs money, and big infrastructures for the mass market cost big money, and so there has to be a big payback, which may look uncertain. BT is not the only telco to agonize over this. But, if a market and industry are really shifting, there comes a point where that investment becomes table stakes – and without it, participation cannot continue. BT has clearly reached that point. As in the stock market, when the most skeptical throw in the towel, you know the market has cleared.
And even Light Reading itself is a minor indicator of fiber's coming of age. The monthly hit rates of two-year-old (and over) reports on FTTx-related topics (see PON & FTTx Update and VDSL2) are still rock solid at a high level – so there is real interest out there on telco fiber access.
"FTTH as a science project – those days are over,” says Floyd Wagoner, director of product marketing for Motorola Inc. (NYSE: MOT)'s access networks business. "The global broadband market today is served by two areas from a telco perspective: DSL and fiber. DSL accounts for 66 percent of the global broadband connections worldwide, and fiber accounts now for 11 percent. I think the shift to fiber has absolutely occurred, and will continue to grow."
The pressure for widespread fiber is mounting as users and governments fret over the threat of a widening gap between the haves and the have-nots. Says Andrew McGrath, commercial director at ntl:Telewest Business :
"The last pan-European broadband speed study found that only six other nations in a league table of 23 European countries have slower speeds than in the U.K. This is down in large part to legacy networks still depending on antiquated copper wiring to transmit high volumes of data. The lag is frustrating enough for consumers, but represents a serious threat to profitability for organizations relying on connectivity. Businesses also need to use media-rich applications to gain competitive edge and drive efficiency. The technical capabilities of next-generation networks are required more than ever to support this shift."
So fixed telecom operators are embracing fiber access. The transition to fiber will take a long time and will probably prove tortuous in many countries, but the landscape has definitely begun to change, and this is going to affect many different players – service providers, vendors, users, regulators, and so on – in possibly unforeseen ways.
That we are talking of an era of transition has implications of its own. In many countries only a minority will have fiber access for quite a time, and maybe some will never get it. How will service providers cope with such a split? What does it mean for service evolution? For regulatory remedies?
Interestingly, fiber may be the first example of such an "era effect" in the technology transformation of the telecom mass market. The three big other technology transformations – mobile, Internet/Web access, and basic broadband – although they had fairly long gestation times before they entered the mass market, were rolled out in large amounts within only a few years once a threshold had been crossed. But fiber looks a much bigger, more difficult, and longer job for many telcos.
So this report tries to take a very provisional look at how some aspects of mass-market fiber access are evolving and how the industry is responding to some of the new issues raised.
PON is merely pandering to the regulators, i.e. the FCC. Nobody in a free market would buy this technology. It's dud technology from the get go. I think of it like ethanol as an alternative fuel - politicians shoving bs down are throats for the sake of special interests. Economic sink holes.
An FTTx solution has to deal with technology churn. The way to do this w/FTTx is to have the customers purchase the equipment and lease the dark fiber. This works for businesses but doesn't really scale to homes. So the reality is FTTx, right now, only makes economic sense to some business parks. (That's why we see folks like GOOG and Marting pitching white space instead of really addressing the hard problem.)
OK... let me clarify what I meant. When I said that the service provider can give customers 100M or 1Gig I meant that if you look on some of the Active Ethernet vendors out there, the same CPE is used for 1gig as well as 100M so for the service provider the only difference would be the price of optics which as the 1Gig optics goes does is almost equal. I didn't refer to LAN elements that the customer can choose to put or not put both in Active and in PON. With that said and with all the cost calculation I've seen for Greenfield I still fully support my statement that for Greenfield Active Ethernet is the better choice.
PON supporters try to hide the limited uplink (ini Active you can get 100M or even 1Gig) and the need for more speed as you need to support more and more HD streams (even with one HD TV when using DVR it means 3 HD streams). Since all the PON vendors understand that they are working for on the next Gen standard. Which will require equipment replacement. You can keep ignoring this aspect or take it into account.
In regards to aggregation cost and opex there are multiple places which have deployed Active Ethernet FTTH (not neccesseraly with Cisco which I agree is a poor choice) which is working for many years now. Their calculation shows that maintaining the network is competitive to PON
In regards to end-user pricing, since the cost of deploying Active Ethernet is competitive to PON (though you can offer a lot more converged services which combine your video voice and data)the offering to the end-user can be competitive or better than PON.
The irony here is that while we're talking about a fiber utopia, Google's Page is cozying up with Chairman Martin, making statements about how the broadcaster's white spaces are the real solution to broaband! Right. The media companies are going to hand over their video advertising revenue streams to GOOG. This is so transparent it's sad.
(Isn't it nice, a public servant doesn't accomplish a thing and then panders a bit to have so insiders hand over wall st. created financial instruments. We've seen this movie before.)
No integrity to be found here nor there. No wonder the world is losing confidence in the U.S. "Change you can believe in" doesn't establish morality and trust. Strength of character comes from somewhere else. That suggests knock off the constant pandering as well as the constant pushing of self interested bs as serving the rest of us.
"If you can find a router that runs 1G for $39 like the one that I have at home that runs 100M, I would really like you to post it."
How about $29? http://www.microcenter.com/single_product_results.phtml?product_id=0238747
But of course this is electrical, not optical. Gigabit optics are pretty expensive compared to 100Mb/s.
To me it is the idea of having active electronics outside in powered air-conditioned boxes has got to have a high opex cost. Air conditioners and optics fail. Power to these boxes goes out. You have to send a guy in a van to fix it. Passive optics is going to sit there working for 100 years. The optics in the CO might fail, but that is much, much cheaper to repair.
SureWest's CTO Bill DeMuth says "The network RTs [remote terminals] are Cisco 4510s (in air-conditioned cabinets, which DeMuth describes as "not ideal, but manageable"), then says "if we ever want to evolve in the future, say, to a PON, it is very easy to just change out the electronics and put splitters back at the remote terminals. "
Evolve from active Ethernet to PON? Interesting choice of words. And they are using GPON in their Kansas city buildouts.
Yes, I'd rather have active Ethernet to my house. Heck, give me Gigabit Ethernet and charge me $9 a month. But realistically, for large buildouts companies are more likely to go for the lowest cost to build and lowest cost to manage.
So again, this means that the numbers are debatable. People are making different choices based on different criteria.
So, to summarize - It is NOT clear by your own admission that Active Ethernet or PON is the better choice. So, I am taking it that you are retracting your statement that Active Ethernet is clearly right in all cases for Greenfield.
However, you have made another mistake. The cost of a 100BT port and the cost of a 1 GBE port are not the same on any box. Thus 100M or 1G is not the same cost. It is a different cost. If you can find a router that runs 1G for $39 like the one that I have at home that runs 100M, I would really like you to post it.
Also, PON will NOT give a max of 60Mb/s. It will give an average of 60Mb/s to a user if ALL users are using maximum bandwidth all the time and no other users in the network are using bandwidth. Now that GigE switch given that it can do 1G per user now has 1G uplink capacity in the Metro OR it is oversubcribed and has an average bandwidth of much less than 1Gbps. The PON is exactly the same. It can burst up to 1Gbps (if you have an ONT with a 1GBE port on it - and they exist).
So, the ACTUAL bandwidth limiter is the uplink from the OLT or the Ethernet Switch. All PON does is require yet another level of sharing and traffic management.
So, you have many incorrect assumptions and statements. I think you should rethink them.
I don't argue, it is clear that each technology has its plusses and minusses. However there are several misconceptions that needs to be broken.
You see very similar prices of CPEs of Active Ethernet and PON, however one will give your customer maxium 60Mbps (and in most cases half if not less) while the other can give the customer 100M or even 1Gig - at the same cost. The biggest expense in FTTH is actually the trenching cost which is completly equal in both technologies. THe fact that one operator showed these numbers doesn't mean that it is the same for all. Many other operators did their own calculation and got to the opposite conclusion - meaning Active Ethernet was more cost effective per customer (and needless to say per Mbps...)
PON vendors already understand that the bandwidth they provide this is why they work on the next gen PON techbology. So a service provider that install GPON today will need to replace it (as the technologies won't interop) in the future, while Active Ethernet give the service provider and its customer more bandwidth with no need to replace the technology or the elements in the network.
In addition the ability to provide business cusotmers different better service than consumers is somthing that can't be done at all with PON
So to summerize calculating price per customer while ignoring the services , bandwidth and QoS that one can offer its customers as wellignoring the amount of time the technology can last is not the right calculation. Even if we assume that in some cases the ROI for the Active is a bit longer (and it is defintly not true to all cases) it won't require re-installation and replacment of equipment for many years to come. While PON will...and at the end of the day installation costs that needs to be re-done again and again are way higher than the cost of the equipment.
PON = 1 laser at the CO per 32 or 64 customers. Active Ethernet - 1 laser at the CO (or remote aggregation point) per customer.
And the point I have already made many times, and all (including BT) have agreed with this that the cost for PON is less than it is for Ethernet. The cost of an OLT port is not 32x the cost of an Ethernet port. Because of that the OLT cost is much less than the cost of the Ethernet switches on a per subscriber basis.
From an ONT standpoint, people do not have optical Ethernet ports on their PCs. So, you have a box at the home. Now, in Japan for EPON and other International spots for GPON, indoor ONTs are going to carriers for under $100.
I can tell you for a fact that Verizon is seeing customer problems (once installation issues which are not technology related) are solved at or below POTS rates on PON. So, make all the arguments you want about OPEX - but it has not shown to be true in the real world.
So, the cost per customer connected (as the other active Ethernet proponent linked the BT analysis) is lower in PON than it is in Active Ethernet. The BT study pegged this cost at 18%. I did not make this study, and the other proponent said that extra 18% was worth it. That is a choice the carrier can make. But it is not obvious that Active Ethernet is the way to go for greenfield. It is a choice and has plusses and minusses.
Well, I'm not optimistic about fiber access. It seems our politicians are about to hand off massive subsidies to the auto industry. I think in the U.S. there are something like 300M cars, about one per person (including the unlicensed and unborn.) Do we really need to do this?
Cars, roads, oil. No change here despite the rhetoric. Is there a progressive anywhere in the house?
The blogs and comments are the opinions only of the writers and do not reflect the views of Light Reading. They are no substitute for your own research and should not be relied upon for trading or any other purpose.
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Digg
Del.icio.us
Reddit
