Slashdot Mirror
Bell Labs Achieves 3.28Tbps Over Fiber
Dave-V writes: "Scientists at Bell Laboratories said they have set a world's record by transmitting 3.28 trillion bits of data per second over 300 kilometers of fiber optic cable. The research arm of Lucent Technologies said it was the industry's first demonstration of long distance, triple-terabit data transmission. Researchers achieved those speeds using Lucent's experimental optic fiber, called TrueWave. Bell Labs scientists said they used three 100-kilometer fiber spans to transmit 40 gigabits over each of the 40 wavelengths of light (colors) in the conventional C-band frequency range and 40 Gbits/s over each of the 42 channels in the long-wave L-band range."
The FoxNews article contains more details. With Iridum about to heat up in the worst way, and landlines jumping in capacity, maybe the future really does hold a fiber-optic link straight into every permanent structure on Earth.
I'm just as excited about this as everyone else is, but when you stop to think about it, what's the point? Let's do the math:
This is a backbone technology, not feeder. Think in terms of a great many nodes feeding into switches. The local switches are connected by 40Gbps connections. Multiple local domains are agregated into the 3Tbps backbone for long haul to the next major city.
If the system is used for a SAN, the inter-city connection would be used to have an offsite mirror for disaster recovery. It would probably serve many customers rather than just one (1 customer needing 3Tbps would be a HUGE customer). Certainly, no single disk drive could move that fast, but consider one 40Gbps channel into a switch serving 60 file servers each with a large RAID.
ISDN was a late '70s, early '80s technology. It wasn't aggressavly marketed, well, ever (by the telcos that is). It wasn't lighlty marketed until late in the '90s. It was very hard to buy in the early '90s (like it was hard for ISPs to buy it, and they were use to talking to telcos then).
The more intresting table would be for when T1s, T3s, fractonal T3s, OC1, OC3, OC12... were actually available from a telco. Not when they were "designed" but when they could be bought. Unfortunitly the closest I can come to putting a date on any of those numbers is "frac T3s in the late '80s", and I'm not even positave about that one.
Bandwidth has been growing a lot lately, but that's unsupprising, research into it has been better funded lately. An intresting issue is what you need to route (or even switch!) data moving that fast. Juniper has nice products, but this is a hell of a lot of bandwidth. Fortunitly (and unfortunitly) it is on a lot of diffrent colors, and you could optically split them and send them to diffrent boxes to route/switch... but that only buys you so much, and it costs a lot too.
1982: 1200 bps
1986: 2400 bps
1991: 9600 bps
1992: 14.4 Kbps
1996: 28.8 Kbps
1998: 50.0 Kbps
2000: 128.0 Kbps(DLS)
_________________________
how many mp3s per second on Napster is this?
-- Count Spatula: The Culinary Vampire "...because my cooking sucks."
According to researchers, the experiment used both DWDM -- a technology that combines multiple wavelengths onto a single fiber -- and distributed Raman amplification -- a technique that allows optical fiber to amplify the signals traveling through it.
This is absolutely not my field, but isn't distributed Raman amplification a way by which the fiber has been 'doped' with some molecule in its structure by which a beam of light gets amplified, so you need less repeaters for a certain distance to carry the same amount of data.
The repeaters are quite cheap when compared to the expense of laying the fiber. Upgrading the max speed of fiber like this is quite awesome to see - if they can keep the rate of bandwidth increase up, the might never have to lay more fiber on their backbones again!
The problem is that most of the fibre that is in the ground now is not capable of amplifying the light so you need more repeaters built into the network to reach these amounts of bandwith.
Use Adsense for Charity
Sorry, but you're incorrect. What one can see happening in Third World countries is that their telecommunications systems are getting more modern then in Western countries. For years they hardly had a telecom structure and now when they are finally implementing it, they choose for fiber instead of copper, because it is actually alot cheaper to install. Vietnam is a good example of that. Moore's law also had its effect on telecommunications.
Use Adsense for Charity
On the other hand, you also have the earlier Bell Labs article; according to that, they managed 160 billion bits on a single wavelength; this, on the other hand, is likely more of a public use of the method. In comparison, however, it doesn't size up; tat 160 billion bits per wavelength, it would have only taken 20 wavelengths to manage this throughput, and I'm reading the article as them having 4 times that many wavelengths. Perhaps, at a certain point, they merely cannot distinguish that much data on cluttered wavelengths yet? Seems like a disappointment, after the hopes of 160 billion bits across 1000 wavelengths. Then again, maybe we'll be seeing Bell Labs breaking yet another record sometime soon.
It is great that they have shown the possibility to send this amount of data over a network. It would basically mean sending the entire contents of all harddisks (2000x) on my universities campusnetwork in about 1 minute. But when are we going to see this technology in service? It seems that not only do we need new repeaters, but also souped up glassfiber. Those are large investments and it may take some time too to get the prototype to become a real world model.
With Iridum about to heat up in the worst way, and landlines jumping in capacity, maybe the future really does hold a fiber-optic link straight into every permanent structure on Earth
My personal opinion is that fiber is definitely the way we are going to go espescially for long distance data transfer. The problem with satellite technology is the lag in the signal and the problem with wireless is that it has too low a bandwith. On the other hand fiber should be able to transmit a signal in 0.2 seconds to any place in the world. So a system where the last mile is covered by wireless and a backbone of fiber seems to be the most plausible way. Interesting little tidbit is that 0.2 seconds is also the maximum lag in a telephone conversation, before people judge it as unnatural.
Use Adsense for Charity