Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Technology

Pure Optical Network Switches 111

Posted by Hemos from the very-fast-ah dept.
richi wrote to us about the all optical switch that was announced from Agilent. The primary reason for the coolness factor is that an optical switch means that an optical signal doesn't need to be converted into electric at the switch, then back to optical.
This discussion has been archived. No new comments can be posted.

Pure Optical Network Switches More

Pure Optical Network Switches

Comments Filter:
  • Optical is the only way to fly, as electricity in copper is achieving its theoretical limits.
  • So does this mean we'll be getting 100% optical CPU's soon?
  • The link in the main story is broken. It looks like this:

    http://slashdot.org/<A HREF=

    Wussup with you people?
    --

  • The Coming of the Fibersphere (Score:5)

    by interiot ( 50685 ) on Tuesday March 07, 2000 @04:11AM (#1220481) Homepage
    See The Coming of the Fibersphere [upenn.edu], a great essay about something that this switch makes possible.
  • The correct link is here [yahoo.com]...
  • I expect the URL should be: Agilent [agilent.com]
    All Optical! I have been waiting for this. Now what about the dark fiber?
    ]{

  • Wow (Score:1)

    by NIVRAM ( 105867 )
    All I have to say is *drool*, someone wanna send me one of these for... St. Patty's day or something?

    --NIVRAM
  • Since the link is broken at the moment, here's a link to the story: Agilent Optical Switch [excite.com]
  • Lucent did this about, what, a year ago? Read here [bell-labs.com]. Nortel has already done this as well. Here's the Nortel link [nortelnetworks.com].

    Optical switching is really cool, don't get me wrong! The slowest link in any big telco network is the switching equipment. How are you going to resolve Nortel's new 2.4Terabit pipes with OC-3 or OC-12 switching technology? Guess what, not happening.

    Have a read about what Lucent and Nortel are doing...very cool stuff.

  • Just look in the page source - like any good slashdotter should:)
  • I was wondering how they were gonna make mirrors that understood where a packet should go, but then I read the article. There's no mirrors. They're using a bubble-maker out of an old cannon printer. So they've made IP-aware bubbles instead. Or maybe it's that they have electronics reading the headers and then creating an optical route? That seems like it would be the weak link in the chain to me, but then, my mirror sometimes doesn't even get the picture of me back to my eyes right, let along reliably switching high-speed data...
  • The press release says the switch will operate by producing bubbles, like in inkjet printers?? Duh???
  • When commanded, a bubble is created at the intersection of the appropriate waveguides and the light is reflected down a vertical path to the switched port. This bubble is formed using the same reliable technology now used in inkjet printers.

    Why does inkjet always get described as being reliable? I've never been happy with inkjet products with the possible exception of my original HP Deskwriter. I can only imagine trying to troubleshoot connectivity problems on one of these things. What if you have a particular circuit that doesn't get used for a while. That ink is going to dry up and never inject! I'm really having a hard time even imagining how they can design "carrier class" equipment on inkjet technology.

  • Re:The only way to fly (Score:2)

    by Anonymous Coward
    I disagree. Copper is cheaper and better for developing countries, where the growth of the Internet is more importnant.

    Copper won't be cheaper once fiber optics becomes the dominant infrastructure in the developed world.

    Also, there is the little problem that copper itself is a fairly valuable metal. Third world peasants have a tendency to dig up any copper wire infrastructure that is not guarded, and sell it on the black market. Fiber optics, being nothing more than glass, has virtually no scrap value.

    I don't see optical switches coming to Mozambique anytime soon.

    That's because you won't be seeing much of anything coming to Mozambique soon. By the time they can afford more infrasture, fiber optics should have become the standard in the developed world.

  • Re:Um, so what? This is nothing new... (Score:3)

    by hattig ( 47930 ) on Tuesday March 07, 2000 @04:30AM (#1220497) Journal
    Marconi did it last year as well. Marconi's could switch 16 different circuits IIRC, but this one makes no claims as to how many it can do.

    Photonics has a big future, it is obviously important, but unless someone makes an optical switch that can switch thousands of different signals, on many different wavelengths, then it isn't the most newsworthy item.

  • It doesn't use mirrors. There's certainly still many unanswered questions here (it's a press release, not a tech doc of any kind), but the news part of this seems to be the fact that it's using inkjet technology to switch the circuits.
  • Well, on Agilent's website, they say they use planar-lightwave circuits (whatever they are) that intersect at cross points. In the cross points, there's a fluid that allows the optical signal to pass through uninhibited. To reroute the signal, they insert bubbles into the cross point. The bubble refracts the signal, sending it to a different circuit. They say the bubbles can be generated and removed "hundreds of times per second". (See Introducing the Agilent Photonic Switching Platform [agilent.com] for more technical details.)

    So, my question is: Have they perfected some way of creating bubbles of the exact same size every single time, and if so, how? Bubbles don't seem (to me) to be something you can regulate by size easily. Anyone with more info, I'd love to be further informed.

    Eruantalon
  • No slashdot effect, just a typo in the link. You can find it here [yahoo.com].

    ----------------------------------------------
  • If you're interested in photonic crystals, there's a pretty good book:

    Photonic Crystals, Molding the Flow of Light
    By Joannopoulos, Meade, & Winn

    It's a bit mathy (Maxwell's equations) but describes how photonic wave guides work 'n such.

    I didn't realize that the technology was anywhere close to commercial level yet, but I'm a bit curious about exactly what role the bubbles are playing. That's a take that I haven't heard before. It seems like if you have to form physical bubbles that your latency will be adversely affected... which was the whole point. I could be dumb though.

  • Re:The only way to fly (Score:2)

    by Anonymous Coward
    Optical is the only way to fly, as electricity in copper is achieving its theoretical limits.

    Agreed optical has the greater potential, but do we know how much more we can squeaze out of copper? We are already doing much more than was once thought practical.

    Also, I hope that when fiber optics finally make it to our homes, it will be a hybrid fiber/copper system, because even when we no longer use copper for bandwidth, it would be nice to have it there for electrical power.

    I don't like the idea of my telephone having to rely on the power company in order to work. Phones now never go down (unless the line is cut) because they can draw their power over the copper phone line. If the phone companies are smart, they will keep this feature, and thus copper should still be around in some form.

  • Sort of make sense, since curved surfaces reflect, and they're easier to create in a machine than planar surfaces. Image of soap bubbles of different sizes appearing to be different colors, depending on their size (curvature) comes to mind. Bubbles are a natural shape - ever try to blow a square bubble? :) I wonder what substance they propose using?

    If they propose to bounce the pulses off of bubble surfaces, they'll distort them. BUT, if they bounce them AGAIN from an identical bubble, they'll (should) be restored. Interesting.

    Then again, it might be a hidden message. What do you need to blow bubbles? Air, preferably hot air. Well, some sort of vapor anyway. Vaporware. :)

    And if the quality of the waveform is as good as the quality of inkjet printouts, we'll have to design a polymorphic iris for the optical receivers, so they could squint into the fiber to make out the signal! ;P
  • Too bad your link doesn't work. This one works though:

    file:///c:\con\con
    (please don't try this without saving your work)

    It did a very effective job at crashing my W98 box... But then it crashes even without your link (I almost had to reboot my computer again after a near-crash experience when trying to post this comment the first time).

    Thank you.
  • The point is that when someone needs a name for their group or company, they call up these naming companies.

    Hewlett-Packard may not have new people, but the name is new. I was just remembering the story that mentioned it.

  • David Copperfield of technology. (Score:3)

    by Yaruar ( 125933 ) on Tuesday March 07, 2000 @04:54AM (#1220510)
    Looks impressive, but is probably just done with mirrors.

    ;-)

  • Re:IP-aware mirrors? (Score:4)

    by Dhericean ( 158757 ) on Tuesday March 07, 2000 @04:56AM (#1220511)
    My reading of the article is that it is a less responsive level of switching. It is the equivalent of a phone exchange or patch panel that allows you to change the connectivity between a set of fibres. Obviously there is some other method of sending information on which pairs to connect together. Also I would guess the mechanism that generates the bubble is electricity based (though I may be wrong).

    I don't see that this particular technology could be used to switch at the packet level simply because bubbles cannot be created or removed fast enough (we're talking moving physical matter so information would be lost in transition). Or maybe the system can only switch a thousand packets a second. Big gaps but boy do they move fast.

    Optical computers require the switching to be instantaneous and driven by light itself (photoreactive or some such). Using this technology in optical computers would be like using relay switches in electrical computers. However relays at least used to be used in MAUs to connect computers together in a network.

    This technology is useful for the infrastructure of optical computing systems but we need the optical equivalent of the valve or transistor (light amplified or switched using only light) to build the actual optical processor.

  • Fully Optical Systems (Score:5)

    by inj ( 160402 ) on Tuesday March 07, 2000 @05:03AM (#1220512)
    This is a rather important development for optical networks. Other than the initial energy source and the receptor at the end of the line, this technology may be the last piece required for a fully optical system.

    The previous important development in optical transmission networks was the optical amplifier. By doped sections of fiber with a high voltage placed across each section, they were able to create an amplifier. The light waves coming into the doped section of fiber would be amplified.

    To briefly describe it, imagine blinking a small flashlight into a window. On the other side of the glass it looks like a car's headlight rather than a small flashlight.

    Ok, this amplifier is cool but why is it important. The amplifier allowed companies to move away from light to electric to light. In fiber optic system, the signal decays as it travels through the fiber. Once it reaches a certain level, it will become unusable. Thus, you have to boost the signal periodically. Traditionally, communication companies had to take the decayed signal, convert it to electric, then you drive another laser with the weak electric signal. This then completes the light to electric to light again, and you now have a usable signal again. (Until you have to boost the signal again) Anyway, this conversion process created a whole new set of problems. By converting it to electric, you introduced noise. After a "amplifying" the signal a few times, your signal was no longer clean buy now included some noise.

    This optical amplifier was important and lead to the FLAG system. If you read Wired a few years ago, they did a really good job of discussing FLAG. FLAG is the "Fiber Optic Link 'round the Globe" and is a _large_ undersea fiber optic cable reaching from UK, to Egypt, to Malaysia, to Japan. This project and other like it, became feasible because of optical amplifiers.

    Optical switches are important for the same reason. It will reduce noise and increase reliability of the system.

    I would not look to see these switches replace the current technology (as mentioned in a previous post). However, I would expect to see these switches start to become incorporated into new systems such as subsea communication, pipeline right-of-way bandwidth, or other cross country communication networks. There already is a big market for piping ESPN to parts of the country. Imagine that you could route the information more like a phone system than a gas pipeline. This could create a whole new market.

  • And some smoke!

    Uwe Wolfgang Radu
  • Hundreds of times a second? Please, we're moving to optical because of the bandwidth. Communications at the speed of light (in glass or plastic). We're talking multi-gigabits per second. And we want to switch packets at that speed with switches that can switch hundreds of times a second? Who can ponder the packet sizes required; megabytes to be sure.

    Sorry, but switches that require the physical motion of sizeable amounts matter don't cut it for optical communications. If they were talking about nanotechnology, that might be another thing. But as it stands, this must be a warm-up for an April Fool's joke.

    Uwe Wolfgang Radu
  • So does this mean we'll be getting 100% optical CPU's soon?
    (Wishing people would RTFA...) From the article [yahoo.com]:
    Light is transmitted across a horizontal path from the input to output port until a switch command is issued. When commanded, a bubble is created at the intersection of the appropriate waveguides and the light is reflected down a vertical path to the switched port. This bubble is formed using the same reliable technology now used in inkjet printers.
    Unless you want to compute at bubble-making speeds (tens of microseconds) instead of at transistor-switching speeds (picoseconds), you wouldn't want to try making an optical CPU with this technology. Oh, it almost certainly takes thousands of times the energy to create a bubble as it does to switch a gate, so you could have a power-hog of a slow processor too. Finally, the bubble creation is probably done thermally, so you have heating and cooling delays even if you don't do a conversion from light to electricity.
    --
  • ...This bubble is formed using the same reliable technology now used in inkjet printers.

    Erm... mybe I've been shopping for printers at the wrong places, but reliability is not what comes to mind when I think about inkjet bubbles...

    And what are these bubbles made of? Where do they go? Will that switch be dripping from used bubble material? Will I have to refill cardridges? ;-)


  • I don't think this is packet switching -- this is circuit switching.

    Example: incoming fiber #5 has many channels multiplexed onto it. The circuit on channels #4567 needs to go out on fiber #2, but on a different channel, say #1234. This allows the switching to happen without converting the signal to electrons!

    richi.

  • Nortel OPTera (Score:3)

    by regen ( 124808 ) on Tuesday March 07, 2000 @05:30AM (#1220520) Homepage Journal
    Well, as a Network Engineering Consultant in a Heavy Nortel Shop, I have had a chance to see the OPTera and Verselar 25000 in action. These devices are Electro/Optical not pure optical and I believe the break through with the Aligent device is that it is purely optical.
  • I was meaning to keep my mouth shut on this, but I've seen it one time too many lately: The editors of /. are obviously not very aware of the differences between "than" and "then". I've also seen "your" instead of "you're", "to" instead of "too".. I'm holding my breath waiting for "there's" instead of "theirs" :)
  • the switch will operate by producing bubbles, like in inkjet printers??
    Yes. Many modern inkjet printers operate by flash-boiling a bit of the ink to create a vapor bubble which forces liquid ink out the orifice. Since there's a large discontinuity in refractive index between liquid and vapor, you can use this to reflect light. You use heating elements just like ink-jet print heads, and you can get switching rates comparable to ink-jet dot rates. (Mind your temperature limits, this thing isn't going to work very well above the critical temperature or below the freezing point of its working fluid! Sub-cooling will not gain you anything in the overclocking race. ;-)

    It strikes me as amusing that this switch is going to be "vapor"-ware even after it ships...
    --

  • Yesterday the link of a large european ISP went down, because they tried to replace the custom bubble-liquid cartridges of their HP optical switches with an el-cheapo no-name product. Other networks stalled when admins tried to refill their switches using a syringe... :-)

    BTW, dont they have to determine where to swicht what? And they sure dont do that optically, so its still the copper and circuits slowing them down, not? Any answers?

  • This is for routing, not packet switching. (Score:3)

    by Christopher Thomas ( 11717 ) on Tuesday March 07, 2000 @05:36AM (#1220525)
    Hundreds of times a second? Please, we're moving to optical because of the bandwidth. Communications at the speed of light (in glass or plastic). We're talking multi-gigabits per second. And we want to switch packets at that speed with switches that can switch hundreds of times a second? Who can ponder the packet sizes required; megabytes to be sure.

    That's not what this is designed for. You'd actually use it for things like this:

    • Routing around damaged backbone nodes.
      If a backbone node goes down, it's not going to go back up a microsecond later. You want to switch _all_ traffic to an alternate route, and then switch it back a few minutes or hours later when the node goes up again.
    • Dynamically adjusting bandwidth for backbone pipes.
      Think of this switch as acting something like a crossbar bus, connecting pipes point-to-point. Need more bandwidth between point a and point b? Allocate an additional pipe connecting them. Not using all bandwidth? Remove a pipe and allocate it to another pair of servers. Load patterns vary over minutes or hours, not microseconds, so this works fine.


    Now, a purely-optical switch that _could_ work on the microsecond or nanosecond level would be very nice; however, a slower switch is still very useful.
  • In that case, this can't be more than an interim solution then. From my humble knowledge of the subject matter, in the long term I believe circuit switching is giving way to packet switching on a grand scale, certainly on the backbones. It simply makes much more efficient use of the pipe. Then again, if they've poured money into this, there must be a market for it, so I'll be still.

    Uwe Wolfgang Radu
  • An optical switch is one thing, but it doesn't do much that a technician swapping cables can't do (only slower). An optical router , now, that would be the biggest enabling technology for an all-optical network. It's also the most difficult.
    --
  • OK, I get it. In that case, this seems like less news than it appears to be, because it applies to a fairly obscure part of the market. A general purpose optical packet switch would seriously be news. I don't see why they don't use TI's quasi-nanotech DLP chip technology for circuit switching like this--it would seem an ideal application, at least superficially.

    Uwe Wolfgang Radu
  • Snip from the article:

    The platform's flexible, modular architecture makes it easy to build optical cross connects (OXCs)

    Seems that you can build router with these.

    Jeff
  • Snip from the article:

    The platform's flexible, modular architecture makes it easy to build optical cross connects (OXCs)

    Seems that you can build router with these.

    You didn't read the whole article, did you? The switching is done with vapor bubbles (presumably in a liquid medium). Forming and removing vapor bubbles takes tens or hundreds of microseconds. At a mere 2 Gb/sec, a 100-K packet goes by in 400 microseconds; its 128-byte IP header flashes past in half a microsecond. Plus, this thing can switch light beams but it has no way to recognize what it's switching (according to the article, "the platform is bit-rate and protocol transparent"); if it can't tell what a header is, it can't possibly route anything.
    --
  • I may be wrong but I think that the correct word here is "then" (after) an not "than" (used for comparison). But I am not a English native speaker...

  • Slow and not cool... (Score:2)

    by Anonymous Coward
    Did I read that right? They route the signal by making bubbles? You got to be kidding!

    Their claim is "pure optical is faster than optical-electronic-optical", but their system is "optical-bubble optical"--very slow.

    I'm not sure just how fast the inkjet guys can make bubbles, but I estimate ~10,000/sec max based on printing rates. Compare than to MEMS mirrors and diffraction gratings which have been around for 10+ years and operate 10-100 times faster!

    Bubbles sure as hell aren't replacing III-V semiconductor electro-optical devices, unless you can make about 10,000,000,000 bubbles/sec...

  • ...and the first transatlantic communications trunk was made out of what? smoke signals?
  • Maybe you'd have to change the way tcp/ip works slightly. When you start a connection with a remote machine you use a new protocol to allocate a frequency and set up your optical route to have the switches set correctly - then for the actual data transmits you have an extremely quick private circuit to the other node. (Which I know is simlilar to the way phone networks work but what the hey)

    Loz

  • Nanovation Technologies [nanovation.com] prototyped an optical switch on 15 April 1999. Further info here [nanovation.com].

    At OFC2000 [osa.org] today, Nanovation displayed new optical switches, splitters and modulators.
    From their press release [nanovation.com]:
    "These switches, splitters and modulators are the first of what will be an extensive offering of integrated photonic components from Nanovation. These products will help businesses and consumers access the full power of all-optical communications in a cost-effective manner," said G. Robert Tatum, president and CEO of Nanovation. "Companies will now be able to build their own customized, optical integrated circuits, thanks to the advanced capabilities of Nanovation components."

    Nanovation's new and revolutionary Nanoshutterä technology outperforms other optical switching products by providing a latching switch mechanism for reduced power, very wideband operation, and the ability to integrate these switches with other functional components. This innovative patent-pending technology combines silicon MEMS switches with a proprietary silica-on-silicon wave-guide process, which will enable Nanovation to offer optical components not only with substantial size and cost advantages, but unparalleled flexibility in optical systems architecture - all on a single integrated device.

    The offerings from the switch family using Nanoshutterä technology include versions of the following:
    Wide band 1X2 optical switch
    Wide band 1X2 optical switch with integrated 5% monitoring taps
    Wide band 2X2 optical switch
    Wide band 2X2 optical switch with integrated 5% monitoring taps
    Wide band 1X4 optical switch
    Wide band 1X8 optical switch
    Wide band 1X16 optical switch

    Nanovation's new offerings for the 1310 nm and 1550 nm telecommunications bands include versions of the following members of the silica-on-silicon wave-guide splitters product family using NanoblockTM technology.
    Single mode 1X8 wide band optical splitter
    Single mode 1X16 wide band optical splitter
    Single mode 2X8 wide band optical splitter
    Single mode 2X16 wide band optical splitter

    The company also demonstrated its 1550 nm high-speed switching technology using Indium Phosphide materials. First components planned around this technology include sub-nanosecond optical switches and high-speed modulators.

    A listing of their product line is here [nanovation.com]. You can download the specs there in PostScript format.

    I have been following this company with some interest since their mention on /. Q1 1999.

  • A few years ago I saw an article about nanotech hinged mirrors. (hmm, I wonder if I read that on slashdot or not)

    They are pretty much mirrors that lay flat and are raised at an angle but they are molecular scale. If I remember properly they weren't very quick. With more research something like this could be used to switch comminications or bits in a computer.

    I think that the most exiting developments will come from quantum computers. Just imaging a computer that used quantum wells for communication. Theoretically this would allow instantanious communication between two points.

    This isn't even going into quantum computers. This is simply using quantum technology to improve standard processors.

    Binder
  • One of the authors mentioned, Joannopoulos @MIT, has a link to Nanovation Technologies, which company I have mentioned earlier.

    Press release [nanovation.com] and up you go.

  • Many years ago there was an article in Scientific American about fiber optic telephone networks. It had a description of a prototype telephone designed by Northern Telecom that used fiber instead of copper. The telephone converted light to electrical energy to power its circuits. I don't remember what they used for a ringer.
  • Don't see what all the fuss is about. I have had light switches in my house for as long as I can remember. My forefathers can remember a time before light switches though.
  • (I know I will get moderated into the ground but who really cares about karma) Why don't you get your head out of your butts? The article has a hosed link and several helpful people type up fixes at the same time. The time stamps are all within a minute or two which means they were all typing up their responses at the same time or they forgot to hit their reload button in the last 30 seconds.

    You have morerate them all down and redundant?? That's the spirit. Let's go moderate the flooders. I really don't care if I read about grits, or trolls or Natalie ocassionally but I can do without a dozen Naked Jackie Chan's in a row. You must really be wound up tight.

    This is supposed to be a place to discuss topics and generally enjoy what we do. Grow up!! Just ignore it if someone cuts up and gets a little sideways. Your moderation is what keeps the blatant ones going trying to see how far they can go. I finally figured that one out. No more posting replies to Don Knotts for me (Bye Frank).

    I was going to post anonomously but decided unsigned letters don't carry much weight. What was I going to do with that karma anyway?

    /End rant Optical switches are a great idea. I know that media converters have the source of way too many problems where I worked at. The fiber connections were nice but the problems came when making transitions between medias.

  • MIT Technology Review did a Fiber to the Home [techreview.com] cover story in their current issue that is very relevant to this thread. Today's announcement that Lucent, Hitachi, and Alcatel [yahoo.com] have agreed to standardize fiber optic tranceivers looks important, too.

    I want a 100Mb/sec fiber pipe to my condo, and I want it now! (grin)

  • Oh. A high speed reconfigurable patch panel sounds a lot easier to implement, and makes more sense to me... That doesn't realy make it a switch in the "router/bridge/switch/hub" kind of switch that I thought it meant though. :)

  • ...and the first transatlantic communications trunk was made out of what? smoke signals?

    I think the person meant to say "immersed in water and without its protective sheath", but then what do I know?

  • I agree that this all optical in that there is no light to electron conversion, but my question is: How is the inkjet like bubble put in the optical channel? Are they using electrical signals there? Even if they are, they are still better off than Nortel, et. al., since they are now using on electrical signal per packet instead of per bit. This still doesn't make complete sense to me... Oh for a technical description instead of PR.


  • > So, my question is: Have they perfected some way of creating bubbles of the exact same size every single
    > time, and if so, how? Bubbles don't seem (to me) to be something you can regulate by size easily.

    Think Inkjet and Bubblejet printers!

  • ferroelectric Liquid Crystals have properties that enable them to be used a optical switches. here [colorado.edu] is what Colorado are doing. My [cam.ac.uk] supervisor reckons that they are going to be very big. And cool.

  • This is an example of retarded Anonymous Cowards. The link points to www.scoobydoo.com.

    YHBT. YHL. HAND.

  • Re:Liquid, Bubbles & Bandwidth (Score:2)

    by Anonymous Coward
    I used to work at a "promotional graphics" company (i.e. junk-mail factory), and I ran a high-speed inkjet printer that could print at 240 dpi while the paper was feeding through at 1100 feet per minute (yes, eleven hundred feet per minute!!!). The print heads used a crystal to vibrate the ink ultra-sonically to produce inkdrops of uniform size which were then propelled onto the paper by electro-static discharge. This was four years ago. I'm sure the ink-bubble technology has been improved dramatically since then.
  • A transister is basicly just a type of switch. We had tube transisters in the 1920's then discreet component transisters 1942, and finally transisters on a silicon chip. Each new generation of transister offered tremendous improvements over the prior. Now we are beginning to hear about the possibility of creating optical transisters.

    For those of you who don't read Star Trek news that means photon based computing! Keep up the good reporting this really is stuff that matters.
  • have a look here [cam.ac.uk] for details of what they've been doing in my department.

    it's weird: the two guys in the photo must be the only researchers in the place without beards. Maybe they made them shave to be more photogenic for the popular press.

  • >The previous important development in optical transmission networks was the optical amplifier. By doped sections of fiber with a high voltage placed across each section, they were able to create an amplifier. The light waves coming into the doped section of fiber would be amplified.

    Only slightly wrong. You don't use a voltage across the fibre section, you shine lots of light down it, 980nm or 1480nm. This excites the erbium ions in the doped fibre and provides gain at 1550nm (the telecomms wavelength).
  • This must explain why HP/Agilent has been agressively recruiting people with strong backgrounds in injet modeling and simulation.
    That seemed suspicious. Now we know at least one big part of what they were scheming. Very clever design, it seems.
  • I don't remember what they used for a ringer.

    Hrm.. flashing lights?

    [ c h a d o k e r e ] [dhs.org]
  • Single Electron Tunneling Transistors use photons to trigger the switch. You can't find much research information about them doing a web search, but they appear to be a possible nextgen transistor.
  • Dig up any copper ..., and sell it in the black market Well, I happend to live in Mexico, and the real problem were people diggin up fiber optics equipment because it reaches a higher price on the black market than copper. I think people tend to steal things which are expensive, copper is just not valuable enough. That was/is the reason why most communications systems in development countries use micro-waves
  • I saw the story on slashdot. They were using micromirrors, and it cut heat output and power consumption by 99%.

    As for the future of optronic computing, I think we're still a long way off. We don't even know what to do with quantum yet. Still, it looks promising, given that I (hopefully) will still be alive in 50 years to this happen.

  • This is a good step forward, but it's not the type of switch most people assume when they see the word 'switch'. It looks like Agilent has the world's first pure optical cross connect, which is a significant development.

    Single fibres are never installed, there are always multiples, so that if one gets damaged/breaks/hit by backhoe, there is another that (hopefully) didn't get damanged, and the equipment switches to the backup. That's the application I see for this Agilent development.

    Transmitters for 2.4 Gbit/S optics are EXPENSIVE, can generate a lot of HEAT, and are fairly LARGE. So if you have multiple redundant connections, you currently need mulitple redundant transmitters! Not a great investment... BUT if you can take 1 transmitter and use a 'pure-optical switch' (better known, probably, as a cross-connect) to connect it to one of N possible fibres, well... it saves money and could improve reliability.

    Not only that, but it's independent of the actual technology on the fibre. So it can switch OC-3's (155 MHz-ish), OC-12's (622 MHz-ish), OC-48's (2.4 GHz-ish), and wave-division multiplexed (multi-wavelength lasers on a single fibre) without caring about it... 'cause it's just a mirror!

    This is a Good Thing (tm). HP might not be the first to do it (I had a college professor that was using microelectronic machines [MEMS] to do something similar), but it seems they might be the first to mass-product it.
  • Does this also mean I could get an ODN link to the Internet? My 56k modem isn't fast enough. -grin-
  • A super broad band net will facilitte two-way
    video in every room of your house, office, public
    building and portable message devices.
    That is the ultimate end of the net.
    Maybe around 2020.
  • Yesterday, Cisco announced [cisco.com]
    it will create "the the first end-to-end, all optical network to deliver Internet access service...." They talk about providing 100x T1 speed for each customer for the price of a T1. Yikes.

  • You could have bought one of these babies in 1997 from the Black Box catalog. I dug an old one out and found it in there..100% optical switch for fiber networks. The US gov is worried about these things because you can't "wiretap" them because there's no wires (obviously).
  • How is the inkjet like bubble put in the optical channel? Are they using electrical signals there?

    The inkjet technology that HP (and now Agilent) uses is a thermal-electric one. You heat the fluid medium with a small resistor. In the print heads, the resistors are spaced a ways behind the print head orifices. Dunno how these are done, but I'd guess they skipped the orifices. If you watch the power, you can have the bubbles collapse shortly after the power pulse is over.

    Bottom line, the control for the switcher is electronic, but the signal path is all optical.

  • Have they perfected some way of creating bubbles of the exact same size every single time, and if so, how?

    First, a reminder. Agilent was split from HP last year when the computer people wanted to get rid of the non-computing sectors. As a result, Agilent got a lot of information that HP uses, including the rather mature Inkjet technology. (I don't recall when it was first invented, but the original Thinkjet printer was introduced around 1983.)

    Even though inkjet printers have evolved quite a bit over the years, the technology behind them is still the same: you have a small resistor on a thin film substrate, in a fluid medium. In the printheads, there's an orifice plate arranged so that a resistor is just below a hole. I don't know if this uses an orifice (suspect not), but assuming the fluid is well characterized, bubble control can be very tight.

    Disclaimer: I don't work for Agilent's Labs, but have read a lot of information about the inkjets. I have built driver chips for inkjets.

  • I'll miss copper, we had many good times together. My first computer was a magnet wrapped with a copper wire (you could call it a 286 too.) Hopefully fiberoptics will get better. Just think, in 50 years we'll find something better then fiber-optic, and everyone will be pissed because we yoink up everything again and replace it. The only thing is the fiber-optic line put up by my telephone company stops 1 telephone pole before my house. It sucks!! now I can't get ADSL (Zoomtown here in cincinati). I hate living out in the country! I might have to share it with my neigbor and run Coax. networking underground to use it.
  • That's a fascinating idea. Hmmm...never thought about that one. I wonder if there's a way to split or copy the wavelengths?

  • It's my understanding that since it is carried over glass and then switched via a mirror there is no way to pick up signals like you can on CRT monitors, and there is no way to put a breakpoint in the line to divert some of the flow to a different computer, like is possible with wire. Therefore, fiber optic ethernet (100base fx I think?) +OC-3 (fiberoptic linkup) +128bit encryption would render a system unbreakable (to the greatest extent humanly possible).

  • Actually, water is very bad for fiber. If I remember correctly from class, its the -OH ion in particular that interferes with and degrades the fiber.

    Most, if not all, outdoor fiber is encased in some sort of shielding to protect it from UV rays, water and rodents (squirrels, mice, rats, etc). Submarine fibers are encased in multiple layers of metal and other materials to form a heavy protective cable. Some possible water hazards -- sharp rocks, anchors, propellors, sharks (yes, sharks!), and obviously the water itself.
  • ARGH!!!
    I dont think it is humanly possible to incorrectly spell 'transistor' more than that many times in one paragraph.
  • If these optical switches are anything like HP's printers, you'll be able to buy 'em really cheap but you'll get ripped off for the ink ;-)

  • By doped sections of fiber with a high voltage placed across each section, they were able to create an amplifier. The light waves coming into the doped section of fiber would be amplified.

    Actually, the doped fiber section is not connected to a high voltage, but to a pump laser, which excites the doped ions in the fiber. Ions (rare earth metals) in an excited state make lasing possible at the signal's wavelength.

    Basically the doped fiber is a laser without a resonator cavity => light coming into the doped section will amplify.

  • Technically what you said makes sense, however some serious security issues are surfacing lately regarding ATM. I'm speaking here without knowing much at all about ATM per se, but it seems to me anytime you connect two pipes together and "let it flow", you're loosing control over the flowing data. I assume circuit switching inherently means that you're not examining the actual data itself. However, in order to implement packet-level security, I don't see any way around working at the individual packet level, which implies packet switching. If the ATM security issues are related to its circuit switching aspects, maybe circuit switching itself is suspect from a security standpoint.

    Uwe Wolfgang Radu
  • Theoretically yes.
    ( I may be full of horse pucky here, it has been a while since I read this )

    In Quantum theory it is possible to have two particles which exist in a sort of dual state.
    Whatever you do to one happens to the other. Example: you have these two particles, you change the spin on one the spin on the other changes also. Instantaneously bacause you actually affected both particles.

    Now, if you use these two particles for communication then the information would actually travel faster than light. Granted this depends a great deal on the feasability of creating the particles (which I believe has been done in a lab), how small you can make the equipment, and the speed at which you can change their state.

    Just may random ramblings.

    Binder
  • Who cares? Agilent advertises a sub 10ms switching system. Lucent's Wavestar Lambdarouter advertises a sub 1ms switching using their microstar MEMS system. I lifted the following off of Lucent's website. I hope they don't mind my pimping their system: "The optical layer bandwidth management used in the WaveStar LambdaRouter is made available thanks to Lucent Technologies MicroStar Technology. MicroStar Technology leverages Bell Labs' patented research in the area of Micro-ElectroMechanical Systems (MEMS) technology. "MicroStar technology is used to attain relatively large switching fabrics with sub-millisecond switching speed and a small product footprint. MicroStar relies on an array of hundreds of electrically configurable microscopic mirrors fabricated on a single substrate to direct light. The switching concept is based on freely moving mirrors being rotated around micro-machined hinges." Lucent's Lambdarouter is already commercially available along with an entire sweet of complementary products. Agilent won't have their product available until late 2000. Sounds like a game of catch-up to me. All of these poor suckers who ran up the price of Agilent stock might want to get out while the getting is good. Check out these two companies websites and tell me who you think will win.

Slashdot Top Deals

Your own mileage may vary.

Close