Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

New Internet2 Land Speed Record 338

SquadBoy writes "An international team set a new record for Internet performance by transferring the equivalent of an entire compact disc's contents across more than 7608 miles (12,272 km) of network in 13 seconds. The rate of 401 megabits per second achieved in transferring 625 megabytes of data from Fairbanks, Alaska to Amsterdam in the Netherlands is over 8000 times greater than the fastest dial-up modem."
This discussion has been archived. No new comments can be posted.

New Internet2 Land Speed Record

Comments Filter:
  • Just in time for Doom 3 :)
  • that's fast (Score:5, Interesting)

    by Kargan ( 250092 ) on Wednesday May 22, 2002 @07:21PM (#3569056) Homepage
    If I'm not mistaken, that's approximately as fast as a 7200rpm ATA/66 drive can transfer data, say, to another partition on the same drive, or what have you.
  • Internet2 (Score:2, Interesting)

    by linuxator ( 529956 )
    Take a notice, that Internet2 project is right now only for universities and big companys... And right now - for testing pourpourses only...

    (sorry for my bad english)
    • Actually, its not big companies. I think its just govt. and universities. Well, with the exception of maybe cisco or M$. Ah well. My university is on the internet2, however we really can't utilize it. Does anyone know if it's on IPv6 or what?
    • by FyRE666 ( 263011 ) on Wednesday May 22, 2002 @07:53PM (#3569264) Homepage

      Take a notice, that Internet2 project is right now only for universities and big companys... And right now - for testing pourpourses only...

      Which examination are the pourpourses taking?

      Good luck to all the pourpourses out there!!
  • by dada21 ( 163177 ) <> on Wednesday May 22, 2002 @07:22PM (#3569063) Homepage Journal
    The RIAA and MPAA have both come out proclaiming Hayes' new 4800 baud modem with MNP5 as the best connection system possible, and are subsidizing the conversion from broadband to these hardware devices with a $50 rebate until the end of the year.

  • by bravehamster ( 44836 ) on Wednesday May 22, 2002 @07:22PM (#3569066) Homepage Journal
    747 chock full o' DVD's can beat that any day of the week. That includes a 4 hour layover in Osaka.

    • by Jon Abbott ( 723 ) on Wednesday May 22, 2002 @07:29PM (#3569104) Homepage
      Andrew Tannenbaum put it best with, "Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway."
      • Andrew Tannenbaum put it best with, "Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway."
        This math is not done by me, but by Bonboard on Everything2 (search "Never underestimate the bandwidth of a station wagon full of quarter-inch tapes"):
        13 Petabytes per second.

        For comparison purposes, this is equivalent to about 650 strands of perfectly saturated, single-mode fiber optic cable.

        This figure will, of course, vary depending on a number of factors. In order to compensate for your own rate of travel and storage media, simply fill in the blanks below to get your tally! It's fun for kids of all ages!
        BW = (( WV / (TW * TL * TH ) ) * TC * WS / WL) , where

        BW = bandwidth in bytes / second
        WV = the volume of your station wagon, in cubic meters
        TW = the width of each individual quarter-inch tape, in meters
        TL = the length of each individual quarter-inch tape, in meters
        TH = the height of each individual quarter-inch tape, in meters
        TC = the capacity of each individual quarter-inch tape, in bytes
        WS = the speed of your station wagon, in meters/sec
        WL = the length of your station wagon, in meters

        This figure assumes average instantaneous bandwidth down the length of the wagon; in reality, I would assume that the bulk of the data transfer would occur in the region nearest the trunk.
        To get my figure, simply plug in: WV = 2.72, TW = 0.054, TL = 0.073, TH = 0.0105, TC = 35.0 * 10 ^ 9, WS = 26.8, WL = 4.75. These numbers are meant to describe a stuffed 2001 Subaru Outback doing 60MPH using 35GiB tapes of this form factor.
    • Can it?
      that's 6,646 cd's worth of stuff in 24 hours.
      Your plane would move more than that.

      So if your destination and source media are CD, the airplane is faster.
      If it's NOT, the airplane is much slower, factoring in the time it will take to read all those cds.
  • by Navius Eurisko ( 322438 ) on Wednesday May 22, 2002 @07:22PM (#3569067)
    Marge: "Does anyone need that much porn?"

    Homer (drooling): "One million times faster...."
  • Darn... (Score:4, Funny)

    by errorlevel ( 415281 ) on Wednesday May 22, 2002 @07:22PM (#3569068) Homepage
    Suddenly my DSL no longer seems fast enough.
  • Ping rate? (Score:4, Insightful)

    by Jon Abbott ( 723 ) on Wednesday May 22, 2002 @07:23PM (#3569073) Homepage
    401 Mb/s is great, but what sort of ping rates were they getting?
    • by edrugtrader ( 442064 ) on Wednesday May 22, 2002 @08:15PM (#3569354) Homepage
      lets just say you've lost before your computer even boots up
  • Netjunk. (Score:2, Interesting)

    by saintlupus ( 227599 )
    The rate of 401 megabits per second achieved in transferring 625 megabytes of data from Fairbanks, Alaska to Amsterdam in the Netherlands is over 8000 times greater than the fastest dial-up modem.

    Must be nice to have a pipe that's not full of SPAM, pop-up ads, Code Red, Nimda, SQLSnake, Gnutella, ARP scans from the braindead fucks at my ISP, AIM crap...

  • Is that why my ping spiked....?
  • by fmaxwell ( 249001 ) on Wednesday May 22, 2002 @07:29PM (#3569102) Homepage Journal
    Some day, I'll have that kind of bandwidth running to my home. And my ISP will still disallow my personal telnet server because of the strain it will put on the network.
  • wow (Score:5, Funny)

    by teslatug ( 543527 ) on Wednesday May 22, 2002 @07:29PM (#3569108)
    Finally finding the pr0n becomes the bottle-neck.

    Pr0n jokes are obligatory for this kind of story, read the manual.
  • Somebody's law says the more space you have the easier it is to fill with junk - be it drivespace or bandwidth. Just like 28.8 used to kick ass, I think that 400 mbps will become slow. How soon though?
    • "Hey cool! We don't needs to bother compressing movies anymore, hey we don't even need to compress 'em to DVD quality! We can just send them at the full whopping 250mbps that they take up!"

      "Oh, hang on, damn this 400mbps connections slow - hell I can only download one mega high-quality movie at a time ot it starts breaking up!"

      or something.
  • by DanThe1Man ( 46872 ) on Wednesday May 22, 2002 @07:30PM (#3569117)
    News of faster internet made HIllary Rosen faint.

    Someone shouted, "quick get some smelling salts".
    Someone else said, "Here, use this sharpie marker."
  • by Anonymous Coward

    ...when they realised it was Britney Spears' latest album, they sent it straight back even faster.

  • by yobbo ( 324595 ) on Wednesday May 22, 2002 @07:33PM (#3569141)
    Amsterdam and Alaska are separated by water, so I don't see how this can be a new land speed record, unless Jesus is involved.
  • Land speed record (Score:2, Informative)

    by Sycophant ( 4279 )
    If I am not mistaken (and I could be, I suck at maths).

    It traveled at about 3,345,350 KM/H, or about 5,352,560 MPH...
    • no......

      (B = bytes, b=bits)

      625MB * 8b/1B = 5000 Mb / 401Mb/sec = 12.5 sec

      401Mb/sec * 1048576Mb/1b = 420478976b/sec

      You CAN'T derive physical speed from that. Bits travel at the speed of light, and bits per second is dependant on the lenght of time each bit lasts, in this case, a fraction of a nanosecond.

      • Well, you have a metric, which is bandwidth multiplied by distance, in this case 4.9Pbit*m/s... you can get the equivalent metric by sticking 35343 CD-ROMs worth of data in the back of a van and driving it at 100km/h down the freeway. Isn't statistics fun? :)
      • Oh yes, each bit much have lasted at least 40ms -- the time it takes light to travel 12272km in a vacuum. In an optical fiber it would take even longer. Now, if you're transmitting at 401Mbit/s, and assuming the speed of light in vacuum, each bit is just about 75cm long, which means it swoshes by an observer (such as the receiver) at about 2.5ns -- not even that is "a fraction of a nanosecond."
  • LOC? (Score:5, Funny)

    by sean23007 ( 143364 ) on Wednesday May 22, 2002 @07:35PM (#3569149) Homepage Journal
    Yeah, but how many Libraries of Congress is that? Until they release their accomplishment in Libraries-of-Congress-per-second, it means absolutely nothing to me, or anyone else. Right?
    • If a LOC is 10 terabytes, then 700 megs (approximately), is about .00007 Libraries of Congress. (Don't get me started about bips and bits).

      The informaton that the LOC is 10 terabytes comes from the Data Powers of Ten [] page. Whether or not this is entirely accurate, I'd be willing to bet that a lot of reporters and such use it as a reference. They're probably good ballpark numbers. To quote a bit from the section of the page that includes the LOC:

      Terabyte (1 000 000 000 000 bytes)
      • 1 Terabyte: An automated tape robot OR All the X-ray films in a large technological hospital OR 50000 trees made into paper and printed OR Daily rate of EOS data (1998)
      • 2 Terabytes: An academic research library OR A cabinet full of Exabyte tapes
      • 10 Terabytes: The printed collection of the US Library of Congress
      • 50 Terabytes: The contents of a large Mass Storage System
      • More accurately (and relevantly), the transfer rate of 401 megabits per second amounts to 0.00005 LOC/s, or 4.33 LOC/day. This is of course assuming the reference LOC size of 10 terabytes (not tebibytes).
  • Debian
    Now if only Debian was the sort of corporate entity that would use this for it's own propaganda purposes! Maybe IBM will run a few ads of people viewing 9000x6750 streaming video on their Debian 7x7 (hmmm, 6x8 head G200 plus another G200!) xinerama display PC!
  • by wbav ( 223901 ) <> on Wednesday May 22, 2002 @07:43PM (#3569198) Homepage Journal
    This article, I feel is more than a little vauge. What did they use to transfer? Was it just over electrical/telephone lines, or did they use optics? What kind of compression was used, and what kind of signal boosters/optical repeaters were used in sending this. All of these items could be used to affect the speed of transfer, and well, the article just doesn't say. I mean in theory, one could build a router from parallel to serial that could take data at 9.6 terabits/sec. How are they actually measuring things? Just the time between there and here? Using full optical lines, wouldn't they be able to set the record at c * the index of refraction of the fiberoptic line? It would just be a matter of putting all the data into one block of light.

    Also the article suggested only one way communication, what happens with error checking and such?
    • by mindstrm ( 20013 ) on Thursday May 23, 2002 @12:20AM (#3570243)
      You are speaking theoretically.

      Yes, you could send data much faster than this over fiber. We do it in labs all the time. HEck, we do it in networks over copper all the time (GigE..)

      This is an actual working network, and they send the data from Alaska to The Netherlands at 400mbps. That is fantastic.

  • transferring the equivalent of an entire compact disc's contents across more than 7608 miles (12,272 km) of network in 13 seconds

    It's probably the DNA codes for a polar worm.
  • by aengblom ( 123492 ) on Wednesday May 22, 2002 @07:50PM (#3569245) Homepage
    is over 8000 times greater than the fastest dial-up modem

    I just LOVE sem-relevant comparisons! ;-). And the fastest car on earth goes 8000 times faster than I crawl! Next time let's compare it to at least DSL!
  • by JFMulder ( 59706 ) on Wednesday May 22, 2002 @07:52PM (#3569252)
    Napster on Internet 2? Damn these guys are quick!!!
  • by Quizme2000 ( 323961 ) on Wednesday May 22, 2002 @07:52PM (#3569254) Homepage Journal
    "Debian Woody in less than two minutes"

    Any woman will tell you waiting more than 90 seconds for a "Debian Woody" is unacceptable. My personal best is nearly 1/3 of that.

    Its a joke..get it? No...RTFM
  • Old news! (Score:5, Funny)

    by chuckcolby ( 170019 ) <chuck.rnoc@net> on Wednesday May 22, 2002 @07:53PM (#3569259) Homepage
    They actually had it ready 2 weeks ago, but were unable to transfer the cd until they used a felt tip marker around the edge.

  • by anzha ( 138288 ) on Wednesday May 22, 2002 @07:55PM (#3569270) Homepage Journal
    Every year there is a competition at the high performance conference (Supercomputing 2001 was this last one). It is entitled the 'Bandwidth Challenge'. This last year, NERSC [] took first place with a 3.3 gigabit/second sustained graphically represented simulation using seaborg [].

    Now, admittedly, it wasn't intercontinental, only from Oakland, Ca to Denver, Co....:D
  • by RumGunner ( 457733 ) on Wednesday May 22, 2002 @07:55PM (#3569271) Homepage
    I don't think anybody at the University of Alaska-Fairbanks even knew this was happening. I think those crazy Dutch just slurped down the image from our local linux mirror (which is maintained by our LUG).

    It seems sort of bad form. The mirror is there for everyone, but bandwidth for the sake of bandwidth....

    And now of course, EVERYONE will start trying to see how fast they can suck down our bandwidth. I bet my internet connection at work is terrible tomorrow.
  • I'd hit my monthly DL limit in under 10 minutes :-( What would I do for the rest of the month?
  • by reality-bytes ( 119275 ) on Wednesday May 22, 2002 @08:07PM (#3569331) Homepage
    Does anybody remember this urban legend?

    There was a IT engineer based in London who was sacked because he couldn't get the ping rate between the London and N.Y.C. corporate offices below 20ms.......His boss didn't see the "speed of light" as a valid excuse!!
  • by Moosifer ( 168884 ) on Wednesday May 22, 2002 @08:08PM (#3569334)
    Internet2 page has some events and workshops [] that look like they might be really good deals. I registered for the IPV6 3 day workshop at the University of Utah for only $100.00 - as long as it doesn't suck, that should be money well spent.
  • .. how fast Al Gore will take credit for inventing the Internet2 also.
    • I'm sick of this joke© It's based on a quote taken wildly out of context©

      IIRC, the quote actually is *about* Internet2, and Al Gore was one of the key figures in passing the bill that sponsored the Internet2 program©

      Gore was one of our more tech-savvy politicians, and we may have killed his presidency run with a dumb joke© ¥In a race that close, you can blame anything
  • Some Calculations (Score:3, Informative)

    by Traa ( 158207 ) on Wednesday May 22, 2002 @08:19PM (#3569371) Homepage Journal
    The transfer rate of the new records calculates as follows: 625MB over 12,272km in 13 seconds = 590000 MB*Km/s = 0.590 TB*Km/s

    When I drive home from work in a few minutes: 125TB (10^15 Synapses, Von Neumann et al.) at 85 Mph during rush hour (yeah, that's me) = 4.7 Tb*Km/s

    The Boing full of DVD's calculates as follows: 4.7GB * 170.5 Cubic meters cargo space / 175 Cubic cm jewel case * 912 Km/h = 662,515Gb * 0.25Km/s = 1160 TB*Km/s
  • by kalgen ( 224492 ) on Wednesday May 22, 2002 @08:30PM (#3569411) Homepage
    The University of Washington has transmitted 1.5Gbps of HDTV across the country. I guess the new thing here is the intercontinental aspect. Here [] for the UW press release.
  • transferring the equivalent of an entire compact disc's contents across more than 7608 miles (12,272 km) of network in 13 seconds

    Fitting that the first ever use of the internet2 is piracy! I bet it was Eminemn's new cd []...

  • From these previous articles, the way that major broadband providers down under managing - Telstra [], Optus [], General [] are capping Internet usage to 3GB per month! This equates roughly to six ISOs. This means that you could max out your traffic allowance in say 6 x 13s = 78 seconds!!!
    You could kiss broadband content eg. Streaming webcasts-music, demos, Telecommuting (ie Voip-video) goodbye!
  • It doesn't matter at all how much faster than the average dialup modem they transfer data so long as there is still a user with a dialup modem at the other end of the transfer. I am not looking to belittle the fabulous work these folks have put in. Merely pointing out that until we solve the "last mile problem" these efforts are largely wasted.
  • All joking aside.. (Score:3, Interesting)

    by SideshowBob ( 82333 ) on Wednesday May 22, 2002 @08:58PM (#3569531)
    pr0n jokes aside :-)

    They really shouldn't be building up expectations in people's minds that "Internet2" is going to make things faster for them.

    These types of stories eventually wind up in the Tech section of the local newspapers etc. and its A Bad Thing TM to build up mis-perceptions.

    Internet2 is not going to solve last mile bandwidth limitations.
  • Complain as much as you want about all the p0rn jokes around here, but there's one thing that is odd... if you read correctly, they do mention it's going FROM alaska TO amsterdam... now... why wouldn't it be the other way around?

    Is this a conspiracy to bring us the latest in p0rn warez? Artic P0rn? We all know how serious universities research labs are, right? :)

  • by Alsee ( 515537 ) on Wednesday May 22, 2002 @09:30PM (#3569653) Homepage
    Assuming a perfect light speed connection and a straight path with no delays en-route...

    The data was flowing at a rate of 2.33 bits per foot which works out to 11 Megabytes in flight at all times. That's one hell of a juggling act.
  • When something like this comes along and everybody gets to show their mad multiplication/division skills?
  • if DoubleClick/etc just stopped with all of the Flash ads, the rest of the existing internet might be that fast. :-)
  • If bandwidth is all you care about, you could beat that with a trebuchet [] and a station wagon full of DVDs.


  • We've already established that FedEx wins on bandwidth, now what about cost per bit.

    For me to get time on the FedEx petabit jumbo jet costs what, $10E-10/bit? Now presume that internet 2 will have a hundred nodes, and will cost ten billion (optimistic on both counts) so about a hundred million per node or about ten million per node per year. So one second costs about 3 cents, and I get 0.4 gig for it presuming there is the demand for full utilization.

    So where is the scientific reason for spending a hundred times more per bit? If it's a big shipment, I can wait for the plane. If it's a small shipment, I can wait for good old internet 1. If it's interactive, I should fly myself to the computer that's doing the crunching or upload the code to my local platform. I have yet to see a legitimate scientific application for this. Maybe there's a futuristic entertainment angle to internet 2, but should NSF be funding ultra-luxury entertainment?

    Internet 2 is a solution looking for a problem.

    Or maybe sequelitis. "The first one was a hit, let's hurry up and get another one just like it out."

    Bah. I don't pay taxes so people can win pointless expensive races. Show me how this helps anything that is remotely in the public interest.
    • shared processors.
      faster government computer processing
      VR from great distances. example: A chemist in alsake, can be doing real interacive moleculer worl in amsterdamn, on the fly with other chemest from around the world.
      Your inability to see what benefit can come from this is your lack of imagination.
      Blue sky research is one of many valuable process need to get things that improve our lives.
  • what speeds would they get if they used ipv4 with public internet ips to transfer data over these wires? why do they attribute this to 'internet2', and not just a fat pipe?
  • Ahem, I use a Quantum dialup connection with instantanious unlimited bandwidth communications to the Illuminati.

    Check with Dr. Erich von Daniken for technical details.
  • My own personal speed record is downloading from the Apache server on the internal drive of my G3 Powerbook (Pismo), via a Cisco 2924M switch, into my DivX-playing Athlon box. It's only half that fast with a 7200RPM Firewire drive, so there is obviously some limit in the FW link. It takes aIEeeee! longer to copy the downloaded file out of the temp directory than it did to download it.

    Before I tried using Apache, I was only getting a third of that (13 Mb/sec). Anyhow, 50 Mbits/sec vs 420 Gbits/sec... you figure it out. Me, I'll just drool for the drive arrays that ran fast enough to keep up with the link.

  • I may be mistaken, but I thought I read about an OC-768 test by Qwest across the continental U.S. earlier on /. - so why is this different? OC-768 is about 40Gb/s, which is a lot faster than 401Mb/s.
  • Damn (Score:2, Funny)

    by kraf ( 450958 )
    With the Internet2 I could download those looped Scorpion King trailers disguised as AOTC 8000 times faster !

  • Any idea what software they used for the transfer? A simple TCP session won't work - the maximum window size of 64KB limits the transfer rate over large speed*delay products. Were they using a multi-session FTP, or customware, or something UDP-based?
  • Why is this news? It is easy to get a gigabit link between any two major cities in the US, if you are willing to pay. I transfer files at gigabit speeds all the time - granted, across a river and not across an ocean, but is it really any different?
  • "...the equivalent of an entire compact disc's contents"

    What? Was the rest marked out with a Sharpie marker?

  • Damn! (Score:2, Funny)

    by Jonny Balls ( 543700 )
    Thats faster than Yo mama chasing after crack

Loose bits sink chips.