Oooooo. Lots of stuff to comment on.
Underground vs. Overhead. Underground placement of telecommunications
assets is advantageous for a multitude of reasons. The obvious one
being ice storms and wind storms. But there are also other very good
reasons. In many jurisdictions, we pay taxes on the poles. These can
range from $1 per pole to $100 per pole per year depending on
location. When you consider the thousands of poles... Underground
placement is tax free unless we are renting conduit. Another reason
for underground placement is that it lasts longer. By the time you get
to 20 years, aerial fiber is pretty well shot. We also run into a
problem with aerial fiber in that it moves. This causes polarization
mode dispersion (PMD) which wreaks havoc with OC-192 and 10Gbe
circuits. Those little triangles on the fiber are supposed to stop
waving in the wind, but they cause cable breaks during ice storms.
The long-term trend for aggregate bandwidth is a doubling every 18
months. Moore's Law also applies to telecommunications and it has so
since the telegraph was mass deployed. There are plateaus along the
way, but bandwidth and what uses it go hand in hand. Morse code,
voice, data, video, Internet... We can blame Netflix, Hulu and Siri
for the latest bump in demand, but if it wasn't that, it would be
something else. This rate of growth has been consistent for over 150
years and I really don't expect it to stop any time soon.
I'll just use a single cell tower as an example. A tower for a single
major carrier was erected in my neighborhood (that I lived in at the
time) in 1999. This was initially connected with a single T1
(approximately 1.5Mbps). About 18 months later a second T1 was added
(3.0Mbps). Another 18 months later two more T1s were added (6.0Mbps).
Sure enough, a year and a half later there were 8 T1s (12Mbps). The
next increase we installed a channelized DS3 (fiber deployment with
OC-3) and they were up to 16 T1s (24Mbps). In 2008 they had the entire
DS3 (about 50Mbps). Before they could even deploy LTE, the tower was
using an entire OC-3 (three DS3s). This tower, right now, is LTE and
has 200Mbps of service going to it. Yes, it is planned for upgrade to
500Mbps in 2014. This is an example of just one tower that happens to
be representative of the typical cell tower we provide backhaul to.
What about MODEMS for business and home use?
1958, 110 bps
1962, 300 bps
1980, 1.2 kbps
1984, 2.4 kbps - 9.6 kpbs (note the plateau then the continuance of
the trendline)
1991, 14.4 kbps
1994, 28.8 kbps
1996, 33.6 kbps
1998, 56 kbps
2000, >128 kbps ISDN
Then through the 2000s we saw ADSL kick in. Guess what? The trendlines
continued. In 2001, most ADSL customers were buying 128 kbps speeds,
but every 18 months the bell curve of new service contracts were for
double the speed. 256 kbps, 512 kbps, 1mbps, 2 mbps, and so on.
Ever wonder what the average consumption of bandwidth is? It all
averages out to about 5%. You use about 5% of the total amount of
bandwidth. Obviously, there are some people who use a whole lot more
and some people who use a whole lot less, but it comes out to about 5%
of whatever it is that everybody has purchased. As available bandwidth
increases, so does the actual usage. Just notice how the CNN website
is feeding video even when you don't select any video. So does
Facebook. Again, we can blame everything on video and gamers, but the
fact is that everybody has a stake in this. Anybody ever speak to
their phone (Siri, Google voice search)? When the iPhone 4S came out,
the average bandwidth per user doubled.
Those big drums of cable that you see might not necessarily be cable.
We almost always use ducts now. The ducts come in all sorts of pretty
colors, but we usually use orange if there is a single duct being
buried. The actual cable containing the fiber varies in diameter
depending on how much sheathing is required. I have a Direct Bury
fiber sample right here that is the diameter of my thumb (widest part)
that contains 256 fibers. As it is a direct bury fiber (doesn't need
ducts or conduits), over half that diameter is the hard plastic cover,
a metal sheath, strain/pull wire and a fabric sleave. Even at that,
the ribbons of fiber inside are loose and move freely.
I was surprised that Jim's fiber splicer was working outdoors like
that. Once gust of wind and he'd have all his stuff blowing around
everywhere and also it's hard enough to keep the glass perfectly clean
for the splice. We would always use the trailer or at least a tent to
do the splicing in.
A modern splicing machine is a pretty cool device and saves a ton of
time. you feed the strands in from either side, it cleaves the ends
just right, aligns the cores and fuses them. It takes far longer to
prepare the cable than it does to do the actual splicing. You can mow
through the splicing pretty quick once things are rolling. We usually
figure an average of 20 minutes per fiber on the old stuff.
--
Ken Norton
ken@xxxxxxxxxxx
http://www.zone-10.com
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|