Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×

New Data Transmission Record — 14 Tbps 193

deejne writes to alert us to a new bandwidth record: Nippon Telegraph and Telephone has announced data transmission at a rate of 14 terabits per second over a single optical fiber. The paper claims the previous record was "about 10 Tbps." In the new experiment, NTT sent data over 160 kilometers (nearly 100 miles) of optical fiber, in 140 channels of 111 Gbps each.
This discussion has been archived. No new comments can be posted.

New Data Transmission Record — 14 Tbps

Comments Filter:
  • by Anonymous Coward on Saturday September 30, 2006 @04:23PM (#16260477)
    And still nothing worth watching.
  • Preparing? (Score:5, Funny)

    by Anonymous Coward on Saturday September 30, 2006 @04:24PM (#16260483)
    vista.windowsupdate.com?
  • by LiquidCoooled ( 634315 ) on Saturday September 30, 2006 @04:25PM (#16260493) Homepage Journal
    I thought it meant 14 ThePirateBays per second...
    • by doofusclam ( 528746 ) <slash@seanyseansean.com> on Saturday September 30, 2006 @04:27PM (#16260513) Homepage
      I thought it meant 14 ThePirateBays per second...


      Given an hour with that link it's exactly what i'd use it for.
      • Re:Misread title (Score:5, Insightful)

        by Hal_Porter ( 817932 ) on Saturday September 30, 2006 @04:49PM (#16260693)
        A link that fast wouldn't help you. There isn't enough seed bandwidth on TPB to give you 14Tbits/sec, nor is there the backbone bandwidth. And you'd need a hell of a RAID subsystem to manage handle writing at 14Tbits/sec sustained.
        • So don't write the data at all. Once you have enough bandwidth for high-def video in real time, there is little if any need to store any data locally; just subscribe to a VM on a mainframe somewhere. If you can send uncompressed high-def video, even computer games are no problem for remote display.
    • Re: (Score:3, Funny)

      "Tbps" could also be misread as "Tbsp" and therefore refer intead to the unit soon to be used for 3D printing (we wish...).
  • Well, I remember back on my 14.4 modem... those text pages loaded like the wind. I was on top of the world... Then those damned pictures started cropping up on websites. Pictures on the internet? Ha! Then came the 56.6k modem which showed those pictures who were boss. No problems. Oh wait, online gaming? File sharing? Cable and DSL save the day. More than adequate... so now this time it seems we got the good speed coming up before the need for it. Its like always being busy all week and never having time to
    • by EnderWigginsXenocide ( 852478 ) on Saturday September 30, 2006 @04:44PM (#16260665) Homepage
      Quote:

      Well, I remember back on my 14.4 modem... those text pages loaded like the wind. I was on top of the world... Then those damned pictures started cropping up on websites. Pictures on the internet? Ha! Then came the 56.6k modem which showed those pictures who were boss. No problems. Oh wait, online gaming?
      File sharing ? Cable and DSL save the day. More than adequate

      Reply:
      I beg to differ. I have [cough] friends that download movi^H^H^H^H^H content from the internet, and some dvd rips^H^H^H^H^H^H^H database files can be larger than 4GB! Even at a good (cheap) DSL line of 1KBPS it still takes quite alot longer to download content than it would take to go to blockbuster^H^H^H^H^H^H^H the office and pick up physical media with the data on it.
    • I routinely peg my download of 4.5 Mbps. So much so, I'm considering paying $15 more for 6.5 Mbps.

      Until consumer grade broadband catches up to hard drive read/write speeds, it's never going to be fast enough.

    • by mnmn ( 145599 )
      I remember 9600 being fast for the BBS servers.

      However the Internet speed has been stagnant here in Canada for the past 5 years at least. Just the standard ADSL and noone ever gets the 8Mbit download speed as advertised.

      I dont think we'll ever reach 14Tbps while Bell is in power.
      • by x2A ( 858210 )
        I think they're planning on using lines of this capacity for backbones, not to-each-user anyway.

    • by kreyg ( 103130 )
      I imagine the academic uses are more of the motivation than consumer interests. The data from a single particle collision experiment can be several gigabytes in size, so accessing a remote database of them makes high bandwidth quite desirable.
  • by tverbeek ( 457094 ) on Saturday September 30, 2006 @04:29PM (#16260531) Homepage
    That's still nothing compared to a semi loaded with DVDs traveling at 70mph.
    • by CrazyJim1 ( 809850 ) on Saturday September 30, 2006 @04:40PM (#16260637) Journal
      This is the internet, not the interstate.
    • by Indy1 ( 99447 )
      or a C-5 loaded with 750 gig hard drives :)
    • by EnderWigginsXenocide ( 852478 ) on Saturday September 30, 2006 @04:46PM (#16260687) Homepage
      The internet isn't a truck you can't just keep dumping things on it and expect it to go. It's a series of tubes and they are getting filled up!
      • Re: (Score:3, Insightful)

        by SeaFox ( 739806 )
        The internet isn't a truck you can't just keep dumping things on it and expect it to go. It's a series of tubes and they are getting filled up!

        You mean like the highways get filled up with semis and traffic slows to a crawl? Yeah, tubes aren't like highways at all...
    • by Soft ( 266615 )

      Let's say that a DVD's box is 15cm by 15cm by 2mm (about 2000 DVDs per cubic meter), and the semi is 20m by 5m by 5m (500 cubic meters). That's one million DVDs, each containing about 8 GB or 64 Gb, so 64 petabits total. Traveling at 100 km/h (60-70 mph), that makes approximately 20 Tbps over a 100 km link.

      (If hard drives are carried instead of DVDs, I guess that number becomes about 100 Tbps.)

      So, a loaded truck is still better than a single fiber link, but not by an order of magnitude. It's not "no

    • ok 14 Tbps = 1.7 TBps = 438 DVD's per second (Assuming 4 gig each)

      The distance traversed is 100 miles, which would take 1.4 hours, at 70MPH.

      There are 3600 seconds in an hour.

      This means that per hour a line can move 1.58 million DVD's

      for a 70 MPH trip this adjusts to 2.25 Million DVD's

      or 225,000 (100 disk spindles) Each Spindle Weighs 4Lbs

      leaving 900,000 lbs or 450 tons..

      That would be a semi with 200 cars loaded on it....

      Now How big of a truck are you drivin....?

      Storm

      • by tempest69 ( 572798 ) on Saturday September 30, 2006 @05:46PM (#16261103) Journal
        or a 70 MPH trip this adjusts to 2.25 Million DVD's or 225,000 (100 disk spindles) Each Spindle Weighs 4Lbs
        I missed by an order of magnitude here... 2.25 M /100 = 22,500 so this moves it down to a 45 ton cargo . Which isnt even close to the heaviest load on the road.

        So while the new line isnt quite nothing compared to a truck, a truck can move more data 100 miles faster than the new link.

        Storm

        • Re: (Score:2, Insightful)

          by SassyDave ( 557868 )
          a truck can move more data 100 miles faster than the new link

          Until you consider loading/unloading time and writing/reading the DVDs, which would add days of latency. I'm assuming that this fiber line has vritually no latency.
      • Yeah, DVD's wouldn't work well, but consider using 750GB HDs. It would only take three of them to hold that much data. Using your math, that would mean ~15,000 hard drives for a 100 mile / 70mph trip. That's around 9 tons. So maybe 4 cars.

        A large truck can be over 40 tons. So you could get up to about 60 tbps.

        Now, consider a fleet of 100 trucks. 6 pbps!
        • The parent poster just fixed his math, and now it seems only .9 tons of 750GB HDs would be need to achieve 14 tbps. So a 40 ton truck could get about 600 tbps ! 100 trucks would be 60 pbps! That is 3,000 LoC/s!
    • Re: (Score:2, Funny)

      by ziggamon2.0 ( 796017 )
      ... or pigeons carrying hard drives...
    • That's still nothing compared to a semi loaded with DVDs traveling at 70mph.


      Now add on the time to burn to the DVDs then read them back.
  • by KG6 ( 1007815 ) on Saturday September 30, 2006 @04:29PM (#16260537)
    and yet I'm still downloading at a measly 300 kbs.
    • Re: (Score:3, Funny)

      by timeOday ( 582209 )
      Ignore what you read in the news! - bandwidth is a PRECIOUS, SCARCE resource that we must carefully ration and provision by the byte. That's why the Internet is in grave peril from network neutrality proposals, there won't be any bandwidth left over for our innovative new business models!
  • Actually, I was at that postdeadline session. I don't have the proceedings handy, but Lucent reported about the same capacity, something like 15 Tbps over 100 or 200 km (and another experiment with a few Tbps over 200 km, if memory serves). The previous record was set by Alcatel in 2002, transmitting 10.2 Tbps over 300 km, and I believe it still stands as the largest capacity*distance. The distance is important; I'm not sure that there haven't already been 100 Tbps transmissions over a few km -- much eas
    • by Soft ( 266615 )
      and another experiment with a few Tbps over 200 km, if memory serves

      Memory serves, fingers don't. I meant a few Tbps over 2000 km.

    • Re: (Score:3, Informative)

      by Iron Condor ( 964856 )

      1) Yes, distance is cruically important in these measurements. There's no points in having gazillions of petabyte data transfer if it can only done from one corner of the lab to the other. Which is why all credible speed-of-information-transfer articles include a number with units of [ (bits / second) * distance].

      2) The record is still held by the transmissions from Voyager II's encounter with Neptune.

  • Cost (Score:4, Insightful)

    by Bender0x7D1 ( 536254 ) on Saturday September 30, 2006 @04:32PM (#16260579)
    I'd like to know what the cost of the required equipment is. We know that hardware has a premium for the newest and fastest and it would be interesting to see what the premium is in this case. Maybe it would be cheaper to run 14 1 Tbps links instead of a single 14 Tbps link. Sure, if I already have the fiber in place, then using it for higher speed would be the way to go. However, if I am in a position where I am about to lay fiber anyway, I don't really care about those costs since I will be paying them anyway.
    • The problem is the switching equiptment. Nobody had built a switch that can go much above 2 terabits per second, making all extra bandwidth useless unless you lob on a whole bunch of extremely fast demultiplexors to split the 14 terabit speeds 8 ways. (Powers of 2 are Good.)

      On the whole, fiber is cheep. Ultra-high-speed multiplexors and demultiplexors are not. A typical bundle of fibers might easily have 128 or 1024 fibers running through it, and the extra quality needed to go from a few terabits to a few t

      • by x2A ( 858210 )
        You wouldn't want a tube between you and a far state, that'd be costly and have other problems. You'd just want a fatter tube between you and the next state, and a fatter tube between them and the third state. Shorter tubes = shorter distances to find any faults in, shorter pipelines that need to be created before they can be put to use etc etc, and plenty of benefits.

        Also as other poster said, you'd be better doing it by geographical and population properties rather than by lines of authority.
  • by Evets ( 629327 ) on Saturday September 30, 2006 @04:33PM (#16260589) Homepage Journal
    While impressive, the feat was accomplished over a single optical fiber using proprietary amplifiers not in production. It certainly is innovative, but it is not an indication of speeds you will see in consumer level services. I see these high-bandwidth paradigms being very useful in the medical industry in the near future - especially for things like transferring high quality MRI images from hospital to hospital with very little delay, or in transferring patient ICU data to a centralized monitoring center - which is currently being done, but super-high bandwidth models open up avenues of information that are not currently available - anything from real-time HI-DEF video from the room, to real-time control of in-room instrumentation.
    • by wfberg ( 24378 ) on Saturday September 30, 2006 @05:08PM (#16260823)
      While impressive, the feat was accomplished over a single optical fiber using proprietary amplifiers not in production. It certainly is innovative, but it is not an indication of speeds you will see in consumer level services.

      That goes without saying, right? It is, after all, a record. People don't usually turn to the Guinness book of world records for guidance on, say, what a realistic number of hotdogs is to consume within 12 minutes.

      Now of course, greater bandwidth is cool and all, but 14 Tbps is obviously impractical for actual use, even in specialist medical imaging applications -- for the simple reason you couldn't fill up your harddrive (or even RAM) as quick as that!
      • That goes without saying, right? It is, after all, a record. People don't usually turn to the Guinness book of world records for guidance on, say, what a realistic number of hotdogs is to consume within 12 minutes.

        Interesting argument, though I'd counter that the Guinness Book is rather pointedly not a research journal, so people don't interpret "guy eats 200 hotdogs in one hour" to mean that there's a large corporation working feverishly to figure out how to make it possible for joe schmoe to do the sa

      • Now of course, greater bandwidth is cool and all, but 14 Tbps is obviously impractical for actual use, even in specialist medical imaging applications -- for the simple reason you couldn't fill up your harddrive (or even RAM) as quick as that!

        Sure, you'll have trouble finding a single application to handle that data rate. But say that I'm a broadcaster now paying to have my 100 SDI video paths (each 270 Mbps) carried across town on dozens of fibers because each fiber can only carry a handful of them.

    • Re: (Score:3, Interesting)

      My title at work is actually 'Radiology Information Systems Manager', and I'm one of the people responsible for sending MRI images from hospital to hospital, handling video streams from telerobotic surgeries, and the like.

      Surprisingly, data demands in the medical environment aren't nearly as high as you might think. We routinely route MRI images from hospital to hospital with infrared and T1 connections. Those MRI images are actually only about 10MB each. We got ourselves a 1Gb/s imaging network at our
    • > While impressive, the feat was accomplished over a single optical fiber using proprietary amplifiers not in production.

      Yeah, I want to use a rocket sled stacked with DVDs, but I can't find a vendor.
    • I bet they could really use some of this on the backbones of the net though. :)
    • Re: (Score:3, Insightful)

      by YrWrstNtmr ( 564987 )
      While impressive, the feat was accomplished over a single optical fiber using proprietary amplifiers not in production. It certainly is innovative, but it is not an indication of speeds you will see in consumer level services.

      Pushing 56k through a POTS line was an experiment once.
      • I remember working on this wacky thing called "twisted pair Ethernet" with this exotic technology called "echo cancellation." Frankly that flimsy stuff has never had the cachet of half-inch thick yellow coax. Plus, you can't tell when someone down the hall has unplugged the cable from his transceiver.
    • While impressive, the feat was accomplished over a single optical fiber using proprietary amplifiers not in production. It certainly is innovative, but it is not an indication of speeds you will see in consumer level services.

      What would a consumer do with 10 terabits per second? The only comprehensible measure of that speed is "one large cabinet of DVDs per second." It might be nice to have a DVD vending machine that could chug along at that clip, but you'd have to feed it a ton of polycarbonate every 5 min
  • Hardware (Score:2, Insightful)

    by elzurawka ( 671029 )
    i wonder what kinda of hardware you need to send a burst of 14 TBps? is it comming from that much ram? harddrives? U must have some good hardware to be able to queue up that much data and burst transfer like that.
  • by Night Goat ( 18437 ) on Saturday September 30, 2006 @04:40PM (#16260633) Homepage Journal
    Good thing I didn't buy that eSATA card I was looking at today. 3Gb/second? What a piece of crap!
  • by NCG_Mike ( 905098 )
    I get a lower ping in Quake? Seriously though, I think half the time it's the servers on the internet that are slow rather than broadband connections. I'm sure this has some real world use, other than publicity, (stock trading) but I can't imagine many companies needed it - except the obvious googles of the world. Backbones are obviously going to be interested but do they shift that volume of data at peak levels?
  • by Have Blue ( 616 ) on Saturday September 30, 2006 @05:17PM (#16260883) Homepage
    How many DVD movies per second was this?

    Also, they failed to provide a conversion from terabyte to Libraries of Congress.
  • I love acronyms (Score:3, Insightful)

    by necro81 ( 917438 ) on Saturday September 30, 2006 @05:21PM (#16260917) Journal
    Our experiment used the carrier suppressed return-to-zero differential quadrature phase shift keying (CSRZ-DQPSK)*1 format and ultra-wide-bandwidth amplifiers.

    Try saying "CSRZ-DQPSK" three times fast! I guess this acronym does serve the purpose of being easier to say than "carrier suppressed return-to-zero differential quadrature phase shift keying," but couldn't they have chosen a snazzy acronym that was hip to say and then worked out what it meant, like NASA?
    • CSRZ-DQPSK CSRZ-DQPSK CSPZ-DQPSK ... I can't even type that string quickly. It'll need a better brand name.
    • Caesar's Duck P-speak? We could abbreviate that to Caesar's Duck, or use Quack-keying as slang.
  • Arthur C. Clarke wrote a short story that had the telephone networks becoming self aware when the network became sufficently complex. It's possible I tell you, the telephone networks just don't have the bandwi.......
    ......
    .....
    NOTHING TO SEE HERE. MOVE ALONG. /eof
    • Ah, so that's where they got the Terminator 3 plot. Of course, Skynet couldn't possibl.... ...... ....
      NO CARRIER
      • by x2A ( 858210 )
        That's why I use birds to transmit messages; they might be slower, but they're far too stupid to re$@\!... NO CARRIER PIGEON

  • by Tamerlan ( 817217 ) on Saturday September 30, 2006 @05:31PM (#16261003) Homepage
    One of the russian computer trading companies easily topped that. The box with 20 400GB HDDs fell from the shelf 2m high. Total data transmission rate was

    20*4*10^11*8/sqrt(2*2/9.8)~=10^14 bps or 100 Tbps

    As you see if you have enough money to burn you may easily scale that number.

  • What is this Tbps they speak of?

    I thought that we had established that the only true way to measure bandwidth was in "Libraries of Congress" expressed as a function of time?

    Can't we just stick to the standards?

    A reminder for those that haven't been paying attention: data size and bandwith is measured in "Libraries of Congress" Size comparisons for large objects is always done in "Volkswagen Beetles"

    Good day.

    • by hey! ( 33014 )
      What is this Tbps they speak of?

      Tablespoons per second. At that rate you can fill a two liter soda bottle with data in just over two minutes!
  • by khafre ( 140356 ) on Saturday September 30, 2006 @06:04PM (#16261247)
    Current routers, like the Cisco CRS-1, use OC-768c/STM-256, which is about 40 Gbit/sec. Right now, there are a couple of camps in the IEEE, ones that want 40 Gbit Ethernet, others that want the factor of 10 increase that Ethernet has normally been associated with. Since there is no 100 Gbit SONET (that I'm aware of at least), these public demonstrations, this one by NTT and another by Lucent, prove that 100 Gbit Ethernet is possible, even for long haul. Some providers like at&t, Yahoo and Google, really need 100 Gbit Ethernet because they produce that much data, or provide 10 Gbit service to customers, and they need to aggregate it somehow.

  • by kd3bj ( 733314 ) on Saturday September 30, 2006 @06:08PM (#16261281) Homepage
    I once threw a box of 120 Gig tapes into a dumpster. I think there were about 200 tapes in the box.
    I admit the distance wasn't far, but the burst rate was 24 TBytes/sec.
  • I could download the latest Gangasta Ho'z Up tHa Butt No. 17... Sweet!
  • by chrisinsocalif ( 984172 ) on Saturday September 30, 2006 @06:27PM (#16261399)
    Someday our kids will look back at us and wonder how the hell we surfed porn so slow.
  • Is it possible to have that bus speed on a motherboard? modern CPUs are not fed data fast enough from memory, one of the reasons being the bus being so slow.
  • This is why you should flood wire fibre to the desktop in any building you're planning to stay for more than a decade. Then you just upgrade the switches and cards, no need to rewire every 3 years.

     

He has not acquired a fortune; the fortune has acquired him. -- Bion

Working...