Dadamotive: Hyperconnectivity Archives

Path dependencies in fibre plant architectures

Posted by Herman | March 10 | 2010 7:28 PM | permalink

The recent report by Analysys Mason for OFCOM ("Fibre capacity limitations in access networks") shows again that there are some important path dependencies in the choice of fibre plant architectures for access networks. The main stream press focuses on the speed projections but there is a lot more to enjoy.

In a presentation in November in the UK ("Fibre to Britain") I have showed these slides to demonstrate the choices.

Unfortunately Analysys Mason limits their view (as seen so often) to IP traffic : the myopic view of pure Internet afficionados forgetting there is more than data packets, like linear RF. They also miss the opportunity of "Lambda to the Home".

Nevertheless they make the same point: architecture choices create path dependencies.

Each FTTH technology has an upgrade path to support higher capacity for the next five years. However, the fact that TDM GPON solutions require customer premises equipment to operate at a much higher bandwidth than that required for each individual customer, may limit how far it can be upgraded in the medium to long term, especially as the customer premises environment can be challenging. For example the temperature at the customer premises can fluctuate, whereas premises designed to host long-haul network equipment are air conditioned, thereby minimising temperature fluctuation.

The first observation is that more evolution of enabling technologies is required for PON generations than for PTP architecture, which in a way, makes PTP a less risky technology to invest in. In addition, due to the greater bandwidth scalability of PTP systems, the expected life cycle of each generation of PTP is likely to be longer than that associated with TDM PON. This is an important result, as it indicates that operators will have a longer time to get return on their investment with a PTP architecture compared to a TDM PON architecture, despite a slightly more expensive initial deployment costs (PTP are currently 19% more expensive to deploy than TDM PON systems, mainly due to the additional fibre, and therefore duct infrastructure required).

I would add: the one architecture and approach that gives you the highest penetration over the lifespan is by far the most interesting one. Unbundling and flexibility of technology increases penetration so it is a factor you should be more interested in than bandwidth scalability.

So from this analysis, we do not believe that there will be any significant barriers in the evolution of FTTH systems capacity. However, we believe the bottlenecks maybe in the upgrade of one generation of FTTH systems to the next, especially if an operator decide to depart from a TDM PON infrastructure to a WDM PON infrastructure. In the UK for example, if, in the medium term, Openreach was to upgrade its GPON network to a WDM PON network, it would need to prepare for it today by allowing for extra space in the splitter site to be accommodate extra equipment (wavelength demultiplexers for WDM PON).

Indeed.

Not a bad report, just missed some obvious points.

Big company

Posted by Herman | March 02 | 2010 12:16 PM | permalink

Why on earth did anyone at Google think this is necessary or effective as a marketing action?

On major trainstations in the Netherlands, boys and girls dressed in an white Google jacket, dispensing small red pencils for the local elections tomorrow.

ACTA and ineffective strategies

Posted by Herman | March 01 | 2010 3:25 PM | permalink

As the ACTA storm is rising (try #ACTA on Twitter) a new report shows what history has taught us many times (as in the Prohibition): don't fight what everybody wants with severe punishments, create a good commercial and competitive alternative instead.

The Dutch blogger Arnould Engelfriet reports about a Dutch study on the reaction of the Dutch public to a potential "three-strikes" law, (rapport-onderzoek-downloadverbod.pdf) done for the Copyright and New Media Foundation

The translated summary follows.

Download Ban will not help industry

A download ban will not affect the purchasing behavior of Internet users.

Such a ban, proposed by the Cabinet, may prove counterproductive to the fight of
illegal music and movies on the Internet. This shows a survey by research company
Multiscope commissioned by the Foundation and New Media Copyright held amongst a
representative fraction of the Dutch population. The foundation advocates for improving the
quality of legal downloads and the removal of copy protection and unskippable intros. This because sales of legal music will be boosted by these measures.

Of those surveyed 63 percent said they would not buy more movies or music in the store as a result of a ban. Ten percent of respondents said to buy even less in the store after a download ban.
One third of the Dutch population sometimes downloads music or movies from the Internet. More than one million Dutch even more than once a week. No less than 75% of the respondents who is downloading is not going to cut back if such a ban was coming. Only when the ban would be linked to continuous monitoring and a shutdown of the Internet connection as a result of a violation, 43% of the respondents would reduce their downloading.

.....

Over sixty percent of the respondents said they would consider it a mistake and unjustified if the ISP would be required to monitor traffic offenders to detect violations. A shutdown of the Internet connection is considered equal to exclusion.

Offering legal alternatives will give little change to the behavior of downloaders:
only 10 percent of respondents would stop illegal downloading.

Although legal offers of music and movies on the Internet should be promoted, the criminalization of the current download practice is a disastrous policy.

The study shows that the meager supply, the copy protection and the poor quality of legal alternatives are the main reason people download.As the Apple iTunes store for music and a recent British study has shown, investment in quality will be much more productive.

Killer app

Posted by Herman | February 23 | 2010 12:13 PM | permalink

This would make me order a FttH connection for my mother. She would definitely be able to use it and keep in touch with everyone. (CES 2010, LG's new Skype embedded TV series).

Fiber in the home

Posted by Herman | February 22 | 2010 4:02 PM | permalink

The next step in FTTH is obvious: FITH (fiber-in-the-home) and just around the corner.

"Genexis' new CPE is the first POF-enabled product to provide a seamless fiber-powered home network. ......With its unlimited bandwidth, our POF-enabled CPE is a logical step for Genexis as it gives customers a solid, reliable, easy to install and very fast in-house infrastructure. "

The design of OptoLock enables the fiber to be cut and terminated to the exact required length on site, enabling even the most novice consumer to quickly and easily terminate the bare optical fiber. This means that OptoLock enables all the advantages of optical fiber to be brought into the home with do-it-yourself simplicity and costs. The benefits of this simple, robust high-speed interface will be significant as high-speed services like 100 Mb IPTV are delivered into the home.

Symmetry: on averages and variation

Posted by Herman | February 22 | 2010 10:48 AM | permalink

Dr. Kenjiro Cho is a renowned researcher of the Internet in Japan. His famous paper on The Impact and Implications of the Growth in Residential User-to-User Traffic was the first to point to the real observed usage of high speed broadband by a large population.

In my words: at one point in time 5% of the population is a heavy user, but its never the same 5 %. People use the capacity casual when needed and expect an immediate response. Just like we use our electricity network, water network, sewer network. They are not built for averages.

Think of sewers pipes from your home, what a mess would it be if they were designed for the average flow and you try to flush your toilet.

At the IETF meeting in Japan last November on "Internet Bandwidth Growth: Dealing with Reality" Dr. Kenjiro Cho talked about his latest findings (Transcript session IETF76 Dealing with bandwidth growth Cho.pdf)

He used some statistical terminology in his talk that needs some explanation, lest people draw the wrong conclusion. Dr. Cho was kind enough to clarify the data for me, which I would like to share. [Update] The interpretation is mine.

The first data point is the gradual shift in average residential data consumption in Japan.

In the period 2005 -2009 the average upload grew by 29 %, the average download however grew by 117 %. The symmetry ratio's (up:down) changed from 1:1 in 2005 and 1:1,7 in 2009. The main reason is the increase in streamed video.

But averages can be misleading (as in the case of your sewer pipe when you flush the toilet a couple of times per day). An indicator of the variation is the difference between the mean (average) value measured and the mode value of the measured distribution. (The mode value is the value that has the highest count in the table of measurements. For instance the row (400,1,1,1,300,1,2,1,350) has a mean value of 117 and a mode value of 1).

The bigger the difference between mode and mean, the more variation there is in the measured values.

The ratio of observed mean versus mode values are 123:1 in 2005 and 93:1 in 2009 for the upload, indicating a very high variability in usage patterns for the upload. Sometimes some people are sending large amounts of data.

For the download the ratio is 14:1 in 2005 and 8,5: 1 in 2009, less variation than observed in the upload and consistent with the observation that a lot of streamed video is viewed.

Dr. Cho's observations show that bandwidth usage in ultra high speed broadband networks is consistent with a casual usage pattern with very high variations, like we use other utilities.

And that averages can be very misleading.

T.Googlii

Posted by Herman | February 11 | 2010 7:06 PM | permalink

Harold Feld has created a hilarious but otherwise quite believable analogy of Google with a certain type a bacteria, T.Gondii:

For those unfamiliar with T. Gondii, it goes through two stages. In stage one, it can exist in any mammal. But for reproduction, it needs to be inside a cat. So T. Gondii alters the behavior of mice. Mice infected with T. Gondii have their brain receptors flipped so that they find the smell of cat urine attractive. When they find a cat, they respond with aggressiveness instead of hiding, rushing out to fight the cat. The cat, of course, is very happy to devour the mouse. It does not even notice the protozoa cavorting inside it. If it did, it would probably say "as long as it keeps bringing me mice, why should I care that it's using me to reproduce."

T. Googlii is quite similar. It needs to get to us to "reproduce" (make money). To do this, it must alter the behavior of the platform providers so that we consumer kitties can provide T. Googlii with what it needs to reproduce. And while this no doubt appears parasitic from the mouse/carrier point of view, it looks much more like commensalism from the perspective of the consumer/cat.

A great explanation of behaviour.

Gigabits to Googlebits per second

Posted by Herman | February 11 | 2010 3:36 PM | permalink

One of the less understood advantages of an optical network (optical fiber) is the fact that the best fiber does nothing but guide light.

The electro-optical equipment on either end of the fiber creates the communication path. One piece of equipment modulates a (laser) lightsource. The modulated light is guided by the fiber, received and demodulated at the other end by another piece of equipment.

Moore's law applies also to electro-optical equipment: you get more bandwidth for the same price every year.

If you have fiber installed the only upgrade is in equipment, no digging required. Since equipment is depreciated in 3 or 5 years, you can have a fast path to higher speeds.

We have just seen the effect of Moore's law in practice.

The presentation in Zeewolde (Netherlands) by Reggefiber and independent ISP's, skilfully twittered by Vincent Everts (see his video of the high quality videoconference with a schoolclass) not only gave us the announcement of a 200/200 MBps symmetrical Internet access service (Euro 134,95 per month triple play).

The real news was that 1 Gbps optical ports have become so cheap that they will be the standard, replacing 100 Mbps ports.

In other words, the standard capacity available per line or per household is 1 Gbps.

It is a commercial decision to supply less, 10 Mbps and up.

In the USA Verizon recently has tested 10-GPON, supplying 10 GBps down and 2,5 Gbps up, to be shared among 32 customers (max.). The nice touch was that they added the 10-GPON to an existing GPON network while it was operational, ending up with 2 independent GPON networks in the same fiber infrastructure. In one home both networks could be used simultaneously without interference.

Spring is early....

Rearguard fight against new life

Posted by Herman | February 11 | 2010 8:42 AM | permalink

Everybody can't wait for the winter to pass into spring. We need green leaves, more sunshine and a bright view of the future.

In telecoms we can observe the new life pushing hard against the winter.

Last week NLKabel released with much ado a report made by Deloitte.

The hired guns shoplifted selectively from a report by Tim Poulus and repeated the well-known mantra of NLKabel:

- The market in the Netherlands is doing great in broadband, why interfere?

- Cable can upgrade almost indefinitely and deliver anything that is asked (and behold, the demand matches the supply)

- There is more than enough competition, so hands off

I don't want to repeat all international reports from Berkman or the OECD showing that we delude ourselves on our broadband position as far as really high speeds go (let alone mobile broadband).

The most interesting dilemma the report writers faced was how to claim on one hand that cable can deliver anything fiber can deliver, and on the other to prove there is plenty of competition for cable (without fiber).

Because if there is no real competition for cable the argument for intervening because there is a market failure is a no-brainer. If cable, which has been proven to be near impossible to open up for competitors is a defacto monopolist in its capabilities there is a pressing need for a viable open alternative, aka open acces dark fiber networks.

They don't succeed.

VDSL is hyped up to 150 MBps capability, which is bogus. The fact that this capability has to be divided between upstream, downstream and IP-TV bandwidth is carefully neglected.

LTE is pushed as the third alternative. Even politicians not owning an Iphone should know by now that wireless data is heavily constrained. You sell a smart phone that actually delivers on the promise of mobile access and kaboom, the networks buckle under the strain.

We need better infrastructure to get ahead.

The best metaphor is T-Fords and highways. The T-Ford gave everybody access to a new exiting form of mobility with great impact on our wealth. But without the extensive investment in new roads and highways we would not have had a market for fast, cheap, safe, luxurious cars that take us fast and reliable to the beaches at our whim.

In broadband we all have the T-Ford, now we need a much better road system to get to the next level.

Google understands this, so we learned yesterday. Get 50.000 to 500.00 users with near unlimited access (1 GBps) and show 2 things:

- the real economics of fibre deployment (high investment upfront, very low marginal costs off adding bandwidth) to politicians

- the creativity developers have once there is enough capacity and market (see the Iphone and the Android market space)

And in Zeewolde (NL) today the commercial delivery of 200/200 MBps over fibre will be announced. We are still trailing Hong Kong (1 GBps), but the gap is closing.

Spring is coming.

Hotel California

Posted by Herman | February 05 | 2010 8:27 AM | permalink

Jonathan Zittrain comments in the Financial Times on the centrally controlled "paradise" of Apple.

The iPhone's hybrid model of centrally controlled outside software is already moving beyond the smart phone. This is the significance of the iPad. It could have been built either like a small Apple Macintosh - open to any outside software - or as a big iPhone, controlled by Apple. Apple went with the latter. Attach a keyboard to it and it could replace a PC entirely - boasting plenty of new apps, but only as Apple deems them worthy.

If Apple is the gatekeeper to a device's uses, the governments of the world need knock on the door of only one office in Cupertino, California - Apple's headquarters - to demand changes to code or content . Users no longer own or control the apps they run - they merely rent them minute by minute.

The Apple logo may be a sign: once you have taken a bite of the apple you can "check out but never leave".

( Yes, I own a Milestone/Droid for exactly this reason).

Recently in Hyperconnectivity Category