Slashdot Mirror
'Slow' Light To Speed Up the Net
JPawlak writes "Researchers believe that it may be possible to increase the speed of the Internet by slowing down certain parts of it by using metamaterials. Metamaterials could be used to replace the bulky and slow electronics that route Internet information, allowing for faster Internet speeds. As data nears its destination, the frequencies must be separated. The light must then be converted into electrical signals, which are stored, routed, and converted back into optical signals. The conversion not only adds significant cost and complexity to the process, but slows down the transmission as well. However, if the light signals could be slowed during the switching process, they would not need to be converted into an electrical signal. 'The ability to slow the light could be a tremendous force for telecoms that is sure to enhance speed and efficiency,' says University of California professor Xiang Zhang."
"The metamaterials work of Professor Xiang Zhang and his team at the University of California at Berkeley is being highlighted in a paper Wednesday in the online version of the journal Nature and in another appearing Friday in the journal Science."
http://www.newsfactor.com/story.xhtml?story_id=61321
http://www.nature.com/nature/journal/vaop/ncurrent/full/nature07247.html
Sort of. At the moment, all routing is done electronically. The electronics doesn't have as much bandwidth as the fibres feeding it due to slow response times of electronics. So whenever three or more fibres join, you have an information bottleneck. It would be wonderful if this routing could be done optically, allowing much higher bandwidth routers, removing the bottleneck.
But a router is a bit like a set of traffic lights - if two packets of information travelling to the same destination arrive the router at the same time, then one of them needs to be delayed while the other one is sent through. This requires some way of slowing pulses of light.
Today, these people are claiming that metamaterials will be useful for slowing light, and would thus be useful in such an optical router. I'm a tad skeptical about this at the moment (not sure what the losses would be), and there are several other challenges that need to be met in order to create an optical router.
All-optical regenerators (i.e. repeaters) have been developed - I'm not sure whether they've been implemented yet or whether there's a bit more work to be done (I'm not sure whether they support multiple channels yet), but it's certainly a much simpler problem that's much closer to being solved than optical switching.
Fibres can handle much more data than the electronics, so we multiplex the output of 64 electronic systems, send all that down a single fibre, demultiplex it (i.e. split it back into 64 separate fibres) and feed these signals to 64 sets of electronics. I'm pretty sure the multiplexing/demultiplexing is fast, but the problem is that you need 64 electronic routers.
Optical switching would allow much higher bandwidth per channel, meaning you can replace the 64 channels by one, and you'd only need one optical switch rather than 64 electronic ones.
Latency vs. bandwidth: ping times vs. download speeds.
The big advantage of an optical switch would not be a decrease in ping times, the advantage is that the switch has far higher bandwidth. Optical components have much faster response times than electronics, so in theory can support many more bits/sec.
I thought one had to use repeaters every once and a while (every few km?) anyway in fiber optics, which AFIAK work by doing just what this is talking about avoiding, translate light into electrical signals back into light.
They have optical amplifiers these days. http://www.pipeinternational.com/index.php?limitstart=70
Fibre optic amplifiers have been all-optical for about ten years now. It's a neat technology; there's a section of erbium-doped fibre that's made to almost, but not quite, lase, by pumping the atoms up with an external laser light source that's at a lower frequency than the signal. The signal then pushes the fibre over the threshold where there's laser-type activity, and more photons come out than went in.
Yep, but if these are just amplifiers then they amplify the noise too, so every now and then you've got to tidy up the signals and remove the noise, "regenerating" the signal. This is currently done electronically and is another bottleneck. I've just asked a colleague and all-optical regenerators are still being developed.
Demultiplexing multiple channels from an optical fibre isn't routing. This technology could speed the mux/demux stuff up tremendously (saving a lot of cable) but you'll still have a bottleneck at the actual routers that need to read and direct individual packets.
One of these days I'm moving to Theory - everything works there
This article is claiming that metamaterials can give slow light. An optical switch needs a slow light component - where electronic switches can delay packets by temporarily storing them in memory, in an optical switch it would be easier to delay packets by slowing them down rather than stopping them outright. You'd want to delay a packet if it was about to collide with another packet. The actual routing would be done by optical logic, which is very hard and not being claimed by this research group.
Slow light can be achieved in other ways, too - e.g. photonic crystals.
What this is all about is finding some way to do short-term optical packet storage in switches. As yet, there's no optical equivalent of RAM. All-optical gates can be built, and simple logic is possible, but there are no good storage elements. So at present there are optical switches (no queueing), but not optical routers. In order to combine packets from multiple input paths to a single output path, some of the packets need to be delayed until there's free time on the output path. Routers have output queues in RAM to do this. The idea here is to find some way to do this without RAM.
Optical delay lines are simple enough; they're just sections of fibre optic. There are designs for pure optical routers which have little delay loops to which packets can be diverted while waiting for free time on an outgoing line. The delay is fixed, so this sort of thing tends to work better if all the packets are the same size, as in ATM. This new material, where propagation speed varies with light frequency, might be useful as a variable-delay storage medium. Maybe.
This is an area of much active work. Several clever ways have been developed to work around the no-RAM problem. Sort of. None of them are really satisfactory, in the sense of being able to build an optical router that does what an electronic router does now. The network backbone has to be designed around the limitations of the optical technology.
(Note that some optical switches are referred to by their vendors as "routers". They're not. Some of them, the ones with MEMS mirrors, for example, are circuit switches, like a classical phone switch.)
Unfortunately life is not that simple, at 10gigabit you get 33bits per meter. That means that a 1500byte frame occupies about 360m, even if you could knock the speed down 90% you would still need 36m of whatever. And that's just so that you can get it all out before it starts coming back in again.
Your math is off. If we move a wave from one medium to another, the frequency will be preserved, not the wavelength. The "width" of a singal in a slow medium will be far lower as well.
Good point, so since speed = wavelength x frequency, my example of a 90% speed drop will also produce a 90% drop in wavelength so the 360m packet will now occupy 3.6m.
However my implied point that you are going to need a lot of this stuff in your router still holds.
Isn't the speed of light supposed to be constant?
No. Speed of light in vacuum is constant.
Ever heard of ÄOEerenkov radiation which is produced when some particles travel faster than light in water? Off topic, but I always find it beautiful to watch.
hilarious
Delaying or buffering the analog light signal is just a teensy part of the process. A typical packet needs to be detected, isolated, have its CRC checked, be inspected, have its addresses twiddled, have the CRC recalculated, and then queued for forwarding. It's gonna be really hard to do these things optically.
In addition most optical delay devices are going to have a strong phase shift over frequency characteristic, a very bad thing.
Methinks the materials folks should stick with what they know and not speculate on the uses.
The problem is not the math, but the percentage. According to http://en.wikipedia.org/wiki/Slow_light
the speed of light has been slowed down millionfold, allowing for a greatly increased data-density
What?
This is using cut-through, not checking the CRC etc. It's just using a vessel with a known return time to "store" the packet while the route is chosen.
I guess technically it's a hybrid of S&F and C-T.