dslreports logo
site
spacer

spacer
 
   
spc
story category
Researchers Aim to Make Fiber 2,000 Times Faster
Without Much of an Increase in Deployment Cost
by Karl Bode 08:52AM Wednesday Nov 07 2012
UK Researchers at Bangor University in north Wales say they're working on a way to make fiber optic cables up to 2,000 times faster -- without much of an increase in cost. According to the BBC, the researchers have already managed to transmit 20 gigabits of data every second over fiber, and are now in the process of a three-year plan to bring the technology to market. While other fiber researchers examine technology that would increase the number of fiber strands, boost laser usage or improve amplification technologies, the Bangor researchers are utilizing Optical Orthogonal Frequency Division Multiplexing (OOFDM) to boost fiber speeds without higher costs -- and with less power. The team says they've already hit speeds of 20 Gbps and believe they'll soon be able to boost that to 40 Gbps.

view:
topics flat nest 

pawpaw

@fluor.com

FTL?

2,000 times faster than the speed of light? So ping times will arrive before they are sent? Sign me up! High speed trading here I come!
pooker314

join:2005-04-12
Brush Prairie, WA

Re: FTL?

said by pawpaw :

2,000 times faster than the speed of light? So ping times will arrive before they are sent? Sign me up! High speed trading here I come!

Indeed. This will totally ruin my Pandora experience. Generally, I like to put on music based on my mood. But with these kinds of speeds, the music will be informing me what my mood will be in the future. I hate services that screw with my understanding of cause and effect.

cablegeek01

join:2003-05-13
USA
kudos:1

1 recommendation

A step back?

Isn't Cisco already transporting @ 25gbps per lambda with the 100Gbps LR4 module on the CRS platform? As far as I know, 100Gbps per lambda is already reachable and fast becoming the norm....

aefstoggaflm
Open Source Fan
Premium
join:2002-03-04
Bethlehem, PA
kudos:7
Reviews:
·PenTeleData
·Verizon Online DSL

Re: A step back?

said by cablegeek01:

Isn't Cisco already transporting @ 25gbps per lambda with the 100Gbps LR4 module on the CRS platform? As far as I know, 100Gbps per lambda is already reachable and fast becoming the norm....

And what about the cost and power?

They are making so that it cheaper and uses less power.
--
Please use the "yellow (IM) envelope" to contact me and please leave the URL intact.
Skippy25

join:2000-09-13
Hazelwood, MO

Re: A step back?

And they are saying they are doing it without increasing fiber strands.

"While other fiber researchers examine technology that would increase the number of fiber strands....."

You can do hundreds of gig if you multiplex it all together. Their point is that they want to be able to increase the effectiveness of one strand.

cablegeek01

join:2003-05-13
USA
kudos:1

Re: A step back?

said by Skippy25:

And they are saying they are doing it without increasing fiber strands.

"While other fiber researchers examine technology that would increase the number of fiber strands....."

You can do hundreds of gig if you multiplex it all together. Their point is that they want to be able to increase the effectiveness of one strand.

That's somewhat erroneous. Most manufacturers are already working towards/doing higher data rates per lambda (fiber wavelength). Multiple lambdas per fiber (CWDM and DWDM) have been around for many, many years. In most cases, you don't lease dark fiber, you lease a wavelength on transport circuits.

Leveraging OFDM to increase the data rate per lambda is an interesting concept, especially if they're revamping their FEC method like DOCSIS 3.1 is planning, but I'd like to learn more about how they plan to do it.

CXM_Splicer
Looking at the bigger picture
Premium
join:2011-08-11
NYC
kudos:2

Re: A step back?

quote:
That's somewhat erroneous. Most manufacturers are already working towards/doing higher data rates per lambda (fiber wavelength). Multiple lambdas per fiber (CWDM and DWDM) have been around for many, many years.

Exactly. One of the projects I am involved in is a Tellabs ring using the 7100 series DWDM multiplexer. It will transport 44 different 10G circuits simultaneously over 2 fibers... the documentation says it will do 88 but we aren't using it in that application.

That is 880 Gb/s (rounded of course) over a single strand.

Regor48

@distributel.net

Re: A step back?

Not sure what's so ground breaking here, 40 or 80 channels depending on vendor (10Gbps each) are already deployed in the real world. Latency can't improve by deploying higher capacity links, more direct routing (less regen nodes, less distance) is the only way to improve pings.

norbert26
Premium
join:2010-08-10
Warwick, RI

1 recommendation

empty your bank account faster then speed of light

After the sticker shock few will be able to afford it. Most take the lower / medium tiers offered now and its plenty for most households . A major business is another matter but they have high enough speeds available now for most.
xenophon

join:2007-09-17

Re: empty your bank account faster then speed of light

It won't target the end user. This type of tech will be deployed on the backend.

Richie

join:1999-08-26
Tinley Park, IL

Faster?

Are they making the fiber faster(i.e. reducing latency) or increasing the bandwidth of the fiber?

LightS
Premium
join:2005-12-17
Greenville, TX

Re: Faster?

As stated in the article,
they are aiming for bandwidth, not latency.

Latency is (for now) limited to the speed of light, and the speed at which everything can be converted (at fiber modules, transceivers, etc)
Paxio
Premium
join:2011-02-23
Santa Clara, CA
kudos:1
They are making the bandwidth broader, not making the latency shorter. In particular, they are making existing fiber in the ground carry more traffic. That is really good news to fiber broadband providers (like us) because it vastly extends the traffic we can carry without needing to replace our expensive infrastructure -- the fibers in the ground and over your head!
Skippy25

join:2000-09-13
Hazelwood, MO

Seriously guys....

The speed of light as we know it is light in a vacuum. Fiber does not run at that speed and can't because it is 1.) Not in a vacuum and 2.) Running across a cable that introduces some resistance.
JPL
Premium
join:2007-04-04
Downingtown, PA
kudos:4

Re: Seriously guys....

said by Skippy25:

The speed of light as we know it is light in a vacuum. Fiber does not run at that speed and can't because it is 1.) Not in a vacuum and 2.) Running across a cable that introduces some resistance.

Vacuum? Per Einstein, that's utterly irrelevent. The speed of light is fixed. Period.
Paxio
Premium
join:2011-02-23
Santa Clara, CA
kudos:1

Re: Seriously guys....

said by JPL:

Vacuum? Per Einstein, that's utterly irrelevent. The speed of light is fixed. Period.

Well... it's not quite that simple!

Glass fibers have an index of refraction of about 1.5, which means the speed of light traveling in a glass fiber is 1.5 times slower than the speed of light in a vacuum. The speed of light in vacuum is though to be an absolute constant, per Einstein et. al.

»en.wikipedia.org/wiki/Refractive_index
prairiesky

join:2008-12-08
canada
kudos:2
said by JPL:

said by Skippy25:

The speed of light as we know it is light in a vacuum. Fiber does not run at that speed and can't because it is 1.) Not in a vacuum and 2.) Running across a cable that introduces some resistance.

Vacuum? Per Einstein, that's utterly irrelevent. The speed of light is fixed. Period.

huh?. It has different speeds through different mediums
Body Count

join:2010-09-11
Columbus, OH
said by JPL:

said by Skippy25:

The speed of light as we know it is light in a vacuum. Fiber does not run at that speed and can't because it is 1.) Not in a vacuum and 2.) Running across a cable that introduces some resistance.

Vacuum? Per Einstein, that's utterly irrelevent. The speed of light is fixed. Period.

the famous speed of light number is actually when in a vacuum. when light travels through transparent objects, it will slow down. Just like when light travels through water it will slow down.. then when it's out of water it will speed back up again. Know your physics before posting.
MaynardKrebs
Heave Steve, for the good of the country
Premium
join:2009-06-17
kudos:4

Don't worry

The big carrier ISP's will also increase your bill by 2,000 times so it's a wash in the end.

antdude
A Ninja Ant
Premium,VIP
join:2001-03-25
United State
kudos:4

What about deploying into new areas?

What about areas that still don't have fiber services like FIOS?

anon423

@comcast.net

Statistics

Are made up 99% of the time? I'd really love to know where the 2,000 number is coming from....

Anyone by Comcast's Metro-E can get 10gbps links over a pair of fibers (20gbps full duplex). So 20gbps is a 2x improvement. Hardly anything newsworthy...