Forgot your password?
typodupeerror
The Internet Networking Technology

Researchers Break Internet Speed Records 140

Posted by Zonk
from the hey-you-yeah-you-using-firefox-pull-over dept.
MosiMosi wrote to let us know about a new development on the Internet2 front. Researchers in Tokyo have advanced the speed of the network, breaking records twice in two days back in December of last year. "On Dec. 30 [researchers] sent data at 7.67 gigabits per second, using standard communications protocols. The next day, using modified protocols, the team broke the record again by sending data over the same 20,000-mile path at 9.08 Gbps. That likely represents the current network's final record because rules require a 10 percent improvement for recognition, a percentage that would bring the next record right at the Internet2's current theoretical limit of 10 Gbps."
This discussion has been archived. No new comments can be posted.

Researchers Break Internet Speed Records

Comments Filter:
  • by msauve (701917) on Thursday April 26, 2007 @03:12PM (#18889303)
    Don't they have redundant paths? Can't they use ECMP? (I'm assuming that the "limit" is referring to 10 Gbps max link speed)
  • But... (Score:5, Funny)

    by i.r.id10t (595143) on Thursday April 26, 2007 @03:13PM (#18889307)
    But can they beat a station wagon full of backup tapes (or DVDs or whatever) yet?
    • Re: (Score:2, Funny)

      by Rude Turnip (49495)
      Clearly, they must have used very sturdy tubes for this project. Therefore, you could take those same tubes and create tunnels to cross the oceans, which would allow a station wagon full of DVDs to drive around the world. Therefore, it will *always* be impossible to beat the bandwidth of a station wagon full of tapes or DVDs, because the station wagon will just rise with the tide.
    • Re: (Score:3, Insightful)

      by ect5150 (700619)
      Across a 20,000-mile path, I'm starting to bet on the network.
      • by Daath (225404) <lp&coder,dk> on Thursday April 26, 2007 @03:55PM (#18890013) Homepage Journal
        *Never* underestimate an Airbus A380-800F. It will carry a 150 tonne payload at 0.85 mach, 6500 miles before refueling. A Hitachi 7K1000 1TB drive weighs in at 700 g. That's around 210,000 TB. Flight at .85 mach will take about 30 hours, let's give them 10 hours for refuelling and maintenance. That's 40 hours. If I'm not mistaken, that's around 60 GB per second. What's that? Around half a TBps?

        Beat THAT Internet2!

        Feel free to correct my "calculations", as they weren't any such thing! :)
        • Airbus wins (Score:5, Insightful)

          by benhocking (724439) <.benjaminhocking. .at. .yahoo.com.> on Thursday April 26, 2007 @04:01PM (#18890127) Homepage Journal
          But the calculations do need correction. :) 210,000 TB in 40 hours = 1,458 GB/s or 1.458 TB/s.
          • Re: (Score:2, Funny)

            by Hossicle (945360)
            During a recent attempt I got at least a transfer rate 2.1 TB/s through a common phone line.

            Unfortunately the data was just a big string of Zeros...

            Does that count? Great compression rate too!
        • by Hennell (1005107)
          Where a 40 hour rate to get 210,000 TB of information sounds good you have a slight flaw. It'd also take 40 hour rate to get 1 byte of information. Which as I download allot more things in the rate of the later I think I'd get bored waiting around...
          ---
          Contronyms: for people who are chuffed by antonyms
          ---
          • Re: (Score:3, Informative)

            by Nullav (1053766)

            I think I'd get bored waiting around...

            You look at a progress bar for entertainment?
            On a more serious note, you'd still get in one chunk, so the initial byte wouldn't matter.

            What we need to look at is Gigabits per dollar.
            Assuming that you were somehow blessed with an ISP that would let you download over a TB/month, and had a 5Mb/s connection (and assuming constant speed), it would take roughly 19.4 days to download.
            Assuming a 30-day month and that your ISP charged $40/month, it would come to $25.86 for that

            • by Havenwar (867124)
              I have a 10Mb/s (both ways) connection with a monthly cost of under $30. This is a step down from my previous ADSL (24Mb/s) when it comes to downlink speed, but a step up of course when it comes to uplink speed. And definitely a step down in price since the previous ADSL cost me around $60 a month.

              However, neither ISP has batted an eyelash at how much I have downloaded. And yes, depending on luck and quality of torrents, a terabyte takes typically less than two weeks to download.

              Conclusion?
              Move to Sweden. I
        • Re: (Score:1, Redundant)

          by DrgnDancer (137700)
          Ahh, but you have to make time to get the data on and off of the tapes. After all this is a computer to computer transfer.
        • by woodhouse (625329) on Thursday April 26, 2007 @04:58PM (#18891017) Homepage
          Not bad, but I'm not sure I'd want to play CS with that kind of ping.
        • Re: (Score:1, Redundant)

          by Firehed (942385)
          The plane might be able to lift 210,000 drives, but my gut tells me that you'd have a hell of a time getting them all to fit in there...
        • Everyone keeps confusing bandwidth with latency. The bandwidth is the amount of time it takes the data to travel its own length. The latency is the time it takes for the data to get from source to destination.

          At 0.85 mach, a A380 travels its own length in about 1/4 second. So the bandwidth of a A380 is 6720 Pbps. You only need 881,832 A380s to maintain that bandwidth over a 20000 mile course. How to get a 150 ton payload onto or off of an A380 in 0.25 seconds is left as an excercise for the network e

          • Re: (Score:3, Funny)

            by alienmole (15522)

            The bandwidth is the amount of time it takes the data to travel its own length.
            Wow, that makes it all so simple, thanks! Now I just have to measure how long this file I have is. I guess I have to print it out and use a ruler, but what font size should I use? This network design stuff is tricky!
          • by Knetzar (698216)
            Bandwidth is the amount of data that can be sent during a set period of time. Per second in this case.
            Latency is how quickly the smallest piece of information takes to be sent.
        • Re: (Score:3, Funny)

          by TimToady (52230)
          That's nothin'. Never underestimate the bandwidth of an Airbus full of station wagons...
        • by Archtech (159117)
          Yeah, but how long does it take to tear down the hardware before flight and reassemble it afterwards? And what is the error rate (mostly due to drives that don't work on arrival)?
      • If we assume 60mph average speed for that trip, than a 20,000 mile trip will take 333 hours and 20 minutes or 1,200,000. At 9 GB/s, the network will have transferred 10,800 TB in that amount of time. Assuming dual-layer blu-ray DVDs, each with 50 GB (0.05 TB) of data, the station wagon will have to carry more than 216,000 DVDs for it to win. If each DVD takes up about 3.6 cubic inches (0.1x6x6) or 0.002 cubic feet, the station wagon will need to carry 432 cubic feet of DVDs.

        I think the network wins this o

    • Re:But... (Score:5, Interesting)

      by presarioD (771260) on Thursday April 26, 2007 @03:28PM (#18889565)
      But can they beat a station wagon full of backup tapes (or DVDs or whatever) yet?

      Hmmm, let's see: Let's have maximum capacity DVD's at 9GB and for the sake of this exercise let's say the station wagon's capacity is 1000 DVDs so we have 9000GB moving around. Let's say the 20,000 mile distance will be covered at top speed (breaking speed limits in all states) at 100miles/h that results in 200 hours of deliverance time so:

      station wagon data speed = 9000 GB / 200 hours = 45 GB / hour = 0.0125 GB / sec = 0.1 Gbit / sec

      Nope the Japanese win!
      • by goddidit (988396)
        It's all about pipelining the station wagons.
      • by toleraen (831634)
        0.1 Gbit/sec is still a whole lot faster than my Comcast connection...where can I sign up for the station wagon?
        • Re: (Score:1, Funny)

          by Anonymous Coward
          Yeah, but the station wagon has MUCH higher latency (or is Comcast that bad?)
          • by toleraen (831634)
            I can deal with the latency...I just can't wait to see the look on my friends faces when a station wagon pulls up to my apartment to deliver 1000 DVDs!
          • Re: (Score:2, Funny)

            by maynard (3337)
            That's just the fake wood paneling causing too much air friction. Rip that stuff out and you'll get 0.10000100 Gbps, no prob.
      • Re:But... (Score:5, Funny)

        by Obyron (615547) on Thursday April 26, 2007 @03:59PM (#18890093)
        Au contraire!

        Your capacity estimate is way, way too low. My DVD test samples can get 15 discs in a space 1"x5"x5" (e.g., 25in^3). There are 1728in^3 in a cubic foot, which translates to about 69 such stacks, for a total of 1035 discs per cubic foot. With its rear seat folded down the 2008 Volkswagen Jetta SportsWagen has 66.9ft^3 of storage space (source [leftlanenews.com]). We'll call it 67ft^3 for the sake of the math, and assume that you've crammed a few discs in the glovebox. This brings us to a total of 69,345 discs in our datawagon. If we use dual layer blu-ray discs at 50gb/disc that comes to 3.07 petabytes (x10^15). I'll use your 200 hour delivery time, which means we have an overall speed of 269.09GB/s (3467250000000000 bytes / 12000 seconds). You can keep your internet2, although I -will- cede that it gets better gas mileage.

        I would like to posit a new theorum: Advances in storage space and vehicle capacity will always increase such that a sufficiently well-fueled station wagon will have faster throughput than the latest advances in network architecture.
        • Re: (Score:2, Informative)

          by maxume (22995)
          If you are going to use throughput it probably gets a bit scary; raw bandwidth, sure, but I can click send faster than you can burn your blu-ray discs. Your throughput is limited to some multiple of the fastest burner money can buy. If you magic 10x blu-ray burners into existence, you would only need a dozen or so to keep up with the network(of course, you have to go way, way faster than the network because you have all that dead time while you bring the discs over to your readers at the other end of the tr
        • Which is great until you factor in error rate/recovery time(1 bad sector on disk = 1 new trip. yes you could burn parity data too, but then you need to copy a lot more data at once).
          Then theres rate it takes to burn and store disks, and get them all back into a computer on the other end (disk read rate + rate to unpack and move N thousand disks).

          Networks already much faster once you factor that in.

          And of course theres the fact that this multi-gigabit link doesn't need 3.07 petabytes to reach maximum speed.
        • What about double sided discs? :P
        • by mattmacf (901678)

          If we use dual layer blu-ray discs at 50gb/disc that comes to 3.07 petabytes (x10^15). I'll use your 200 hour delivery time, which means we have an overall speed of 269.09GB/s (3467250000000000 bytes / 12000 seconds). You can keep your internet2, although I -will- cede that it gets better gas mileage.

          You sure have a funny definition of "hour." Last I checked, there were 3,600 seconds to an hour, not 60. Redoing your calculations with a value of 720,000 seconds gives us roughly 4.48GB/s or nearly 36Gbps.

          • by Obyron (615547)
            Actually you're right, I screwed up. I knew that somewhere in the midst of all that math I'd make some kind of mistake. To mitigate the damage, I offer the traditional sacrifice of a pop culture quote:

            "Ok! Ok! I must have, I must have put a decimal point in the wrong place or something. Shit! I always do that. I always mess up some mundane detail."
        • Better yet use 2GB micro SD cards
        • by grommit (97148)
          15 discs in 1"x5"x5" ? Those are some extremely thin cases you have for those DVDs.

          Remember, we are talking about data that we'd like to be able to read once the station wagon gets to its destination.
      • Re:But... (Score:4, Informative)

        by SETIGuy (33768) on Thursday April 26, 2007 @05:06PM (#18891185) Homepage

        Hmmm, let's see: Let's have maximum capacity DVD's at 9GB and for the sake of this exercise let's say the station wagon's capacity is 1000 DVDs so we have 9000GB moving around. Let's say the 20,000 mile distance will be covered at top speed (breaking speed limits in all states) at 100miles/h that results in 200 hours of deliverance time so:

        station wagon data speed = 9000 GB / 200 hours = 45 GB / hour = 0.0125 GB / sec = 0.1 Gbit / sec

        You are mixing up latency with bandwidth. The latency (round trip time) of the connection here is 400 hours. The bandwidth (i.e. data rate) is the amount of data divided by the time it takes for the data to travel its own length.

        At 100 mph, a station wagon will travel its length in 0.14 seconds. So the bandwidth of a stationwagon packed with 9000 GB of data is about 550 Tbps.

        Given a train of station wagons running at 100mph, you could sustain that. Of course with 1440000000 ms ping times, I wouldn't try playing Battlefield 2 over that connection.

        Seriously, the distinction is important. If you included transit time when calculating bandwidth, the theoretical maximum bandwidth for a 12,000 bit packet on a 20,000 mile path would be 112 kbps.

    • Re: (Score:3, Informative)

      by Waffle Iron (339739)
      Let's see... A DVD has 3.76e10 bits. If you drove 20,000 miles at 70mph, that would take 1.03e6 seconds. So each DVD in the wagon would give you about 37Kb/s bandwidth. So you'd need about 248,000 DVDs in the car. My little postal scale here says that a DVD in a paper sleeve weighs in at 20g, so you'd need almost 5000kg of DVDs, which is probably too much for a station wagon. You could probably manage the task with BluRay or HD-DVD, though.
      • by Rolgar (556636)
        And this doesn't include the time to get the data from the drive to the network (burning the disks), or get them in the transport medium (load the car), and then unload and put each disk in the drive and remove the data at the other end. Each of those data transfers alone would be slower than this rate of copy. You'd have to seriously parallel copy the data, and get a much faster transport, like a jet to compete with physical transport over this distance. Of course, nothing but data center hardware would
        • Which brings to mind an excellent question: Why burn the disks at all? Why not just disconnect the drive and put that in your station wagon?
      • by markh1967 (315861)
        Don't forget that you also need to add the time taken to write and read those 248,000 DVDs as well. "Insert DVD #148,256 and press any key to continue..."
    • by Lost Race (681080)

      How many DVDs can you carry in a station wagon? 50000? That would be about 400 TB of data. At 9 Gbps they could push about 80 TB per day. They're pushing that data 20000 miles, which is farther than a station wagon can go in 5 days, so it's Internet2 FTW. Even for short trips I think the station wagon would lose once you add in the media transfer time, unless of course data on DVD was what you wanted anyway.

      (My calculations suggest you'd hit the weight limit of the station wagon before filling it up w

      • Re: (Score:3, Insightful)

        by drinkypoo (153816)

        Is there some more mass-efficient storage medium?)

        My guess for the best today is MicroSD. It would be horrendously expensive, but you can get 2GB MicroSD cards. You'd have to amass a lot of MicroSD cards to have the same mass as a CD and it takes only five of them to out-store a dual layer DVD.

        It would take some 25 of them to equal a Blu-Ray disc. Not sure which would win that competition.

    • If you take into the account the time it takes to burn all those DVDs, then yes I think the storage media in a vehicle method was beaten a long time ago.
  • almost but not quite (Score:3, Informative)

    by treeves (963993) on Thursday April 26, 2007 @03:13PM (#18889327) Homepage Journal
    9.08 * 1.1 = 9.988
  • by Glowing Fish (155236) on Thursday April 26, 2007 @03:16PM (#18889387) Homepage
    So with this newer, faster internet, when your staff sends you an e-Mail at 10 AM Friday, you don't have to wait over the weekend to get it?
  • Why is there a limit, surely they can just build wider pipes?
  • Improvement (Score:3, Funny)

    by pizzach (1011925) <pizzach.gmail@com> on Thursday April 26, 2007 @03:25PM (#18889507) Homepage

    On Dec. 30 [researchers] sent data at 7.67 gigabits per second, using standard communications protocols
    Yes, the internet seems to be getting faster bit by bit.

    Ha ha ha *snort* I beat myself up.
  • tubes? (Score:4, Funny)

    by TinBromide (921574) on Thursday April 26, 2007 @03:25PM (#18889509)
    Maybe if they moved from a series of tubes to parallel tubes, they'd get a higher current flow...
  • by zymano (581466) on Thursday April 26, 2007 @03:25PM (#18889513)
    A backbone not owned by the phone companies would reduce prices. An alternative that that doesn't rely on the robberbaron phone and cable companies for the last mile(wimax?).

    Something that allows for video like Iptv would be big.

    It would be more disruptive than the current net because then you could attend classes from home.

    This would be great for the economy too.
    • by drinkypoo (153816)
      If you could just actually use multicast reliably on the internet, you could save an absolute assload of bandwidth, especially with intelligent caching, and/or short start delays so that you only have to distribute a maximum of (length of film / length of start delay) streams.
    • by ckdake (577698)
      So who is going to pay for this alternative? Some of the costs include: -land right of way (railroad companies became telco companies and have right of way between major cites which lets them bury the fiber where they want) (ever priced buying rights to a 10 foot wide strip of land from Atlanta to Dallas?) -burying the fiber -routers that go at these speeds (Ever priced a router that will do a few 10GbE interfaces at wire speed?) -power/cooling/squarefeet/etc in datacenters -people do do and manage all of
  • by smooth wombat (796938) on Thursday April 26, 2007 @03:26PM (#18889529) Homepage Journal
    Marge: "Does anyone need that much porn?"

    Homer (drooling): "One million times faster...."
  • Gee I'm impressed... (Score:3, Interesting)

    by spydum (828400) on Thursday April 26, 2007 @03:28PM (#18889563)
    I'd be more impressed if they DIDN'T modify the TCP stack, and used the PUBLIC Internet. Internet2 is far from a real production network. I'm sure if I ran 40,000 miles of fiber and interconnected two idle routers and modified my TCP stack to handle massive window sizes and other tweaks, I could get nearly the full line rate, at twice the distance.
    • Re: (Score:3, Funny)

      by powerpants (1030280)

      I'm sure if I ran 40,000 miles of fiber and interconnected two idle routers and modified my TCP stack to handle massive window sizes and other tweaks, I could get nearly the full line rate, at twice the distance.
      And if you had, we'd be talking about it.
  • by Rob T Firefly (844560) on Thursday April 26, 2007 @03:34PM (#18889673) Homepage Journal
    Internet2 has just gone even faster, breaking the speed of light.

    An email has just been sent to a researcher on ARPANET in 1972, who unfortunately doesn't know what "v1@gr@" is or why he would want to "enlarge pens" with it.
  • by smitty97 (995791) on Thursday April 26, 2007 @03:35PM (#18889697)
    from TFA:

    With the 10-fold increase, a high-quality version of the movie "The Matrix" could be sent in a few seconds rather than half a minute...

    Efforts to make a high quality version of "The Matrix Revolutions" have not succeeded in any time frame.
    • by Fozzyuw (950608)

      With the 10-fold increase, a high-quality version of the movie "The Matrix" could be sent in a few seconds rather than half a minute...

      So... it's going a few seconds faster than a few seconds?

  • by Anon-Admin (443764) on Thursday April 26, 2007 @03:39PM (#18889757) Homepage Journal
    There 9.06Gbps is a speed record???

    Ummm, OC-192 is 9.6Gbps I think they are a little shy of the speed record. Maybe I missed something.
    • IIRC they're talking about a goodput [wikipedia.org] speed record, not about line rates.
    • Re:New Speed Record? (Score:5, Informative)

      by tomstdenis (446163) <tomstdenis@@@gmail...com> on Thursday April 26, 2007 @04:31PM (#18890599) Homepage
      As I understand it, this is over one link. OC-192 is actually a series of OC-48 links bonded together.

      Heck you can get yourself a nice 10gbit/sec line with 10 1gbit lines, ooh la lah

      Tom
      • it's a single logical link. Perhaps you're confused because the STS hierarchy packs 4 OC-48's into an OC-192, just like there are 24 DS-0s in a DS-1.

        If one is willing to consider multiple links running on a single physical one (i.e. DWDM fiber), 72 x 10 Gbps is possible. If multiple physical links are allowed, then the limit becomes financial/practical.
    • by Zenzilla (793153)
      oc-192 is link bundled. I'm assuming this is over a single link.
    • Re: (Score:3, Informative)

      by ziegast (168305)
      Ummm, OC-192 is 9.6Gbps I think they are a little shy of the speed record. Maybe I missed something.

      Within a data center or a metro area, it's commercially viable to pump tens of gigabits per second of bits from point A to point B using many parallel fiber circuits between the two locations. What makes the Internet2 land speed record (http://www.internet2.edu/lsr/) interesting is adding distance to the problem by multiplying the speed times the distance. The unit of measurement they use is "terabit-meters
  • Does this mean we won't be getting these bi-weekly updates on how Professor Wingwang from Xyzzy University has sent data at ridiculously high speeds over specialised networks using specialised hardware and specialised protocols?

    It's interesting the first time you hear that somebody has sent data at 346363GiB/s or whatever, but there's only so many times you can nod and think "how nice for them" until you start wondering why you're not hearing anything about what's being done to prevent the incapacitation of
    • by biocute (936687)
      Higher speed will be the cure. Just like standard government, the most effective way to fight traffic congestion is to build more and wider roads.

      Imagine what the internet will be like where all spams only counted for 0.01% of the total bandwidth? "They simply cannot breed spammers fast enough to saturate the lines."
      • by Strilanc (1077197)
        I think you mean: Imagine the internet when spam consists of high-def videos involving enlargement.
    • It's really not that specialized. Pretty much every large university in Louisiana is hooked up to this same network, and it's not as if we're like the cutting edge of technology here. I'm not talking "a few big iron systems in a a secluded data center on the campuses of the large universities in Louisiana" I had access to Internet2 on my regular work PC as a junior systems administrator at Tulane, 5 years ago. Made downloading distro .isos much faster (granted the LAN only did 100Mbps, but if that's the
  • What the fuck is this Internet2 thing anyway? Some kind of big truck?
  • This is great! Does this means that my personal internet will be okay even if you put enormous amounts of material, enormous amounts of material in your personal internet? (Weak attempt at Sen. Stevens joke)
  • According to my calc's, the fastest I can send one bit 20,000 miles is in 107 milliseconds. Now how do these yahoo's come up with 9 Gb/s??
  • by rmelton (165795) on Thursday April 26, 2007 @06:13PM (#18892171) Homepage
    With simple assumptions and google calculator

    c / 9.08e9 bits per second =
        the speed of light / (9.08e9 (bits per second)) = 0.264134324 m / Byte

    20000 miles / (c / 9.08e9 bits per second) =
      (20 000 miles) / (c / (9.08e9 (bits per second))) = 116.212843 megabytes

    So bytes are 26 centimeters long, and the network holds 116MB in transit.
  • Why is this "10G" even news? 10 Gigabit (OC192) Has been around since at least 1999. In fact, engineers & scientists already have functioning proto-types of 100 Gigabit over fiber (basically DWDM - multiple colours of 10 Gigabit streams multiplexed).

    The IEEE expects the standard to be ratified in mid 2008 for the fiber version & copper (CAT8?) to come out within a couple of years after that (late 2009 or 2010).

    Siemens achieves 111 Gigabits over 2,400 kilometers
    http://presszoom.com/story_127837.html [presszoom.com]

    B

"Pull the trigger and you're garbage." -- Lady Blue

Working...