Forgot your password?
typodupeerror
Networking Science Technology

First 500 Terabytes Transmitted via LHCGlobal Grid 244

Posted by timothy
from the yeah-but-most-of-it-was-zeroes dept.
neutron_p writes "When the LHC Computer Grid starts operating in 2007, it will be the most data-intensive physics instrument on the planet. Today eight major computing centers successfully completed a challenge to sustain a continuous data flow of 600 megabytes per second on average for 10 days from CERN in Geneva, Switzerland to seven sites in Europe and the US. The total amount of data transmitted during this challenge -- 500 terabytes -- would take about 250 years to download using a typical 512 kilobit per second household broadband connection."
This discussion has been archived. No new comments can be posted.

First 500 Terabytes Transmitted via LHCGlobal Grid

Comments Filter:
  • by cyngus (753668) on Monday April 25, 2005 @03:40PM (#12339782)
    ...this network be able to handle Longhorn SP1?
    • Ha, wonder if the RIAA is going bezerk as they always do, similar to the internet2 controvorsy. "ONE MUSIC CD UNCOMPRESSED IN ONE SECOND" "ONE DVD UNCOMPRESSED IN EIGHT SECONDS"
    • 600 megabytes per second for 10 days -- that's one hell of an mp3 collection.
  • by KinkifyTheNation (823618) on Monday April 25, 2005 @03:40PM (#12339783) Journal
    Can it handle a slashdotting?
  • Great! (Score:5, Insightful)

    by Evanisincontrol (830057) on Monday April 25, 2005 @03:41PM (#12339788)
    Now we don't have to wait around for our porn!
  • by no soup for you (607826) <jesse@wolgamott.gmail@com> on Monday April 25, 2005 @03:43PM (#12339814) Homepage
    //insert perfunctory comment about library of congress here

    On a side note, I tried to find out what the real data size of the LOC is, but I could not.
    • Here you are good sir

      http://64.233.179.104/search?q=cache:asL7GGh_JsI J: en.wikipedia.org/wiki/Terabyte+library+of+congress +in+terabytes&hl=en&client=firefox-a
      • by repvik (96666)
        Apparently, the LoC has shrunk. It now consists of the following lines:

        Server Error
        The server encountered an error and could not complete your request.

        If the problem persists, please mail error@google.com and mention this error message and the query that caused it.

        Of course, that's because of /.'s lame space-in-url-idiocy. Anyways, the LoC is actually approx. 20TB.

  • by prisoner-of-enigma (535770) on Monday April 25, 2005 @03:44PM (#12339822) Homepage
    ...a box full of DLT, LTO, or AIT tapes. With FedEx at my side, I can have several hundred terabytes sent almost anywhere on the planet in 24 hours.

    Of course, the latency for this gargantuan data pipeline is a bit on the high side...
    • by dave-tx (684169) <df19808+slashdot&gmail,com> on Monday April 25, 2005 @03:48PM (#12339870)
      I'm playing Quake III via FedEx, but the prices are killing me.
    • Not really. (Score:3, Insightful)

      by Space cowboy (13680) *

      More to the point, the time it would take to get the data onto and off the tapes is left out of your argument. The bandwidth of a truck full of tapes is an old argument, but they're just so damn slow at both endpoints, they're not that useful after all :-(

      When the data arrives through a network pipe, it's on disk ready to be crunched through whatever program you're running...

      8 or 9 years ago, I used to work in the post-production industry in Soho, London. There's a network called 'Sohonet' where lots of t
      • Re:Not really. (Score:4, Interesting)

        by Stween (322349) on Monday April 25, 2005 @04:18PM (#12340170)
        When the data arrives through a network pipe, it's on disk ready to be crunched through whatever program you're running...

        600 Megs a second. I'd be interested in seeing what sort of disk technology can handle that level of throughput. They must have some amount of buffering going on, hand in hand with the bonus that they're probably able to just stream the data to arrays of disks without really being too concerned about placement (I'm assuming the data transfer is essentially a sequential stream of data, not sodding great numbers of small files, of course).
        • 600 Megs a second. I'd be interested in seeing what sort of disk technology can handle that level of throughput.

          Maybe they're just piped to /dev/null. It's not you can't measure throughput without saving the data.

          • That's true, yes; I was just picking up on Space cowboy's implying that disks are uber-fast, when in fact the polar opposite is the case :)
        • by jd (1658)
          You have 600,000,000 hard drives in a striped RAID array. Then you only have to store 1 byte per second.
        • It was 600 MB/sec, not necessarily one experiment producing 600 MB/sec. Divide and conquer onto multiple RAIDs if you're using linux boxes. Problem solved.

          Or use a high-end SGI box or Sun box with multiple SCSI channels bonded into one filesystem with xlv (or the sun equivalent). Each channel goes to a different RAID. Problem solved.

          Or ... (there are loads of ways to do it, just by dividing the bandwidth up into manageable chunks). Problem solved.

          Simon

    • Haha, I'd hate to have the job of transferring the data to tape and then shipping it simultaneously to 200 places. Good luck keeping up.

      From TFA:
      "When the LHC starts operating in 2007, it will be the most data-intensive physics instrument on the planet, producing more than 1500 megabytes of data every second for over a decade."

      "Scientists working at over two hundred other computing facilities...will access the data via the Grid."
    • ...a box full of DLT, LTO, or AIT tapes.

      No it doesn't. Sure, the part of the transfer than FedEx does for you take under 24 hours, but how long does it take to write the data from the source systems onto the tapes? How long does it take to read the data off the tapes and onto the source systems?

      • Using a large multi drive tape library, probably less than 24 hours. However the fundamental flaw is assuming you have to write *all* the tapes before you ship, and wait for *all* the tapes to arrive before you start the restore.
    • but can even move faster than light on a bicycle?
  • Dark Fibre (Score:5, Interesting)

    by superpulpsicle (533373) on Monday April 25, 2005 @03:44PM (#12339824)
    At least things can transfer alot faster within US, if we actually lit the dark fibre underground. We planted so many during the .com eras, yet so many are still unlit due to unwillingness to the hire more techies for maintainance.

    Well going outside the US is a different story. I really don't know how we connect to Europe etc.

    • Re:Dark Fibre (Score:3, Interesting)

      by Capt'n Hector (650760)
      Come on. The actual fiber is dirt cheap. So what if there's miles and miles of it underground, you still need to hook your little section into the larger network, buy a whole bunch of hardware, AND "hire more techies for maintainance." If it were economically feasible and lucrative, they'd be at it right now. But they're not, seeing as the demand for ultra-high bandwith pipes lie with (a) pirates who don't want to pay for stuff anyway, and (b) CERN. And wuddya know, CERN already has their own ultra-hig
    • Re:Dark Fibre (Score:5, Interesting)

      by jeffmeden (135043) on Monday April 25, 2005 @04:15PM (#12340149) Homepage Journal
      Huh? The cost of laying the fiber pales, as you partly mentioned, when compared to the cost of fiber transceiving equipment. There are tons of 'dark' copper lines as well, that could sustain much greater data throughput than they currently do, but its all about how the data is handled at either end. The same goes for intercontinental runs. You think they are dropping cables one at a time? There is enough bundled up capacity for a long time, but it's 'dark' until it's cost effective to put the gear at either end to use it.
      • One word... eBay! Your source for grey market metro switching hardware.
        In our metro area, prices for business bandwidth (via fiber) are out of sync with consumer grade circuits. Telcove is helping correct the monopoly that Cox Business (was Fibernet) had, to a degree.

    • Dark Fibre

      The Metamucil of choice by all Lord Siths

    • Re:Dark Fibre (Score:2, Interesting)

      by rnxrx (813533)
      It actually isn't about the amount of fiber that matters any more. To an ever increasing degree these strands are being lit with DWDM equipment. The few OC192's these folks are transmitting is a small portion of what's possible. Even without going into particularly exotic or ridiculously expensive equipment it's possible to push 80Gbps+ (e.g. 10 gigabytes per second - or a bit more than 16x the throughput mentioned in the article) out of a pair of strands. Upgrade to the bigger units (or wait a few mont
      • Hence the massive use of outsourced tech support and other jobs but that has a good side for Us as well. Shorty we won't have to work local any more. Your programmers could work just off the beaches of Thailand or Spain, living in a place that's dirt cheap compared to Tokyo or Manhattan and have all the connectivity they need. Very 'Snowcrash' of us.
  • rr (Score:3, Interesting)

    by robpoe (578975) on Monday April 25, 2005 @03:45PM (#12339836)
    I dont know about you .. but my Road Runner is 5mbit/sec .. not 512k. That's only 25 years!

    But seriously. What do you transfer then? I mean, how many Libraries of Congress do you need sitting around on disk.

    • Re:rr (Score:2, Interesting)

      by wed128 (722152)
      This isn't really for consumer use, it's for scientists that need to transfer data from very sensitive, high-resolution, hi frequency instruments. The kind that haven't been invented yet, but will put out that much data. someday. maybe.

      Eh, nevermind...it's a pissing contest.
    • Re:rr (Score:5, Insightful)

      by slavemowgli (585321) * on Monday April 25, 2005 @03:51PM (#12339896) Homepage
      You'd be surprised at the amounts of data captured during experiments in high-energy physics - and keep in mind that this was 500 TB in a *week*, which is longer than you usually want to wait for your data transfer to complete.
      • ...which is why they should probably REDUCE the data on-site, instead of sending it around the world on a massive grid. Isn't bandwidth generally more expensive than computing power these days?
        • They *do* reduce it on-site. That's why they "only" need to transfer 500TB/week. You'd be amazed by the amount of data that gets taken, and thrown out, at a modern accelerator experiment. But I'll leave that for an experimentalist to talk about. (I'm just a lowly theorist.)
      • Re:rr (Score:5, Funny)

        by morcheeba (260908) on Monday April 25, 2005 @04:18PM (#12340172) Journal
        Keep in mind that was 500 TB in a *metric week* -- 10 days. Those crazy physicists!
      • It's expensive, and time consuming and would require every house to upgrade to more expensive hardware. In all reality, there isn't enough demand for super-high bandwidth except for people who need to download large amounts of data. Presently, most of those people are those who like to download large programs - generally programs that they should be buying instead of pirating.

        Yes I would like to have fibre running to my house. And I might notice the increase in speed when playing CounterStrike, downl
    • Re:rr (Score:3, Interesting)

      your 5mbit downstream pales in comparison to comcasts/ool's 10, or fios's 25. But none of that downstream will help you send files, which is what the 512k was referencing (your rr is probably 316 or somewhere around there)

      As for what to send, Physics info works great, as thats what the story is about. For residential use, its all about p2p.
      When 10/10 is standard(VERY capped fiber) expect a nice p2p mounted filesystem, where instead of the traditional p2p process (search, trim results, download file, wait,
  • 42 (Score:5, Funny)

    by ARRRLovin (807926) on Monday April 25, 2005 @03:46PM (#12339846)
    And then they shut the thing down.
  • Cost (Score:5, Funny)

    by aerozeppl (849113) on Monday April 25, 2005 @03:46PM (#12339852)
    Thats great and all but none of us will be on anything like that for years. If Time Warner had that here they would charge one child a month. You would need 12 wives just to cover your internet bill.
  • At last! (Score:5, Funny)

    by ArAgost (853804) on Monday April 25, 2005 @03:46PM (#12339853) Homepage
    The perfect solution to connect my beowulf clusters!
  • by 4nd3r5 (732488) on Monday April 25, 2005 @03:48PM (#12339869) Journal
    Ive been looking all over for it.
  • BROADER band? (Score:3, Interesting)

    by dukeluke (712001) * <dukeluke16@@@hotmail...com> on Monday April 25, 2005 @03:49PM (#12339881) Journal
    Interesting to note, the Internet2 just had a string of lawsuits pertaining to students using the service for illegal filesharing.

    Will this allow you to fileshare so fast that no one can even track it?? Now that would be interesting!

    Seriously though, after reading the article and the miscellaneous links. The numbers were astounding! In comparison to my own broadband, I can get 5 or 6 gigs downloaded in a VERY good day at most. Whereas this network enabled traffic of up to 50 terabytes a DAY! Woot woot! When can I hook up for it?
    • Not really. If your computer is fast enough to d/l the data...someone elses computer is fast enough to get it too. The only thing is - since your d/l time will be shorter it will mean you are on the net a bit less.

    • Seriously though, after reading the article and the miscellaneous links. The numbers were astounding! In comparison to my own broadband, I can get 5 or 6 gigs downloaded in a VERY good day at most. Whereas this network enabled traffic of up to 50 terabytes a DAY! Woot woot! When can I hook up for it?


      Where can I buy cheap, and I mean CHEAP, hard disk drives to save all this good stuff? I can't afford Pricewatch prices at that data rate.
  • RIAA (Score:5, Funny)

    by rdurell (827253) on Monday April 25, 2005 @03:50PM (#12339887)
    Only 10 days? I guess the RIAA sent cease and desist letters.
  • 512 kb? (Score:3, Interesting)

    by Samus (1382) on Monday April 25, 2005 @03:51PM (#12339904) Journal
    Is that 512kb typical household broadband speed upload or download? I guess for upload that makes sense since most broad band connections are not symmetrical. Download is a different story. I have about 3.5 on a dsl and that is fairly typical for the cable guys as well.
    • Depends on where you are. In the UK, the standard base package for ADSL is 256up/512down, with varying download bandwidths into the Mbps.
  • 640 MB/sec (Score:5, Funny)

    by scovetta (632629) on Monday April 25, 2005 @03:52PM (#12339914) Homepage
    "640K ought to be enough for anybody." --Bill Gates, 1981

    "640MB/sec ought to be enough for anybody." --Me, /., 2005
    • "640K ought to be enough for anybody." --Bill Gates, 1981

      "640MB/sec ought to be enough for anybody." --Me, /., 2005


      "It's never enough...for anybody." -- Me, /., 2005
      • "640K ought to be enough for anybody." --Bill Gates, 1981

        "640MB/sec ought to be enough for anybody." --Me, /., 2005

        "It's never enough...for anybody." -- Me, /., 2005


        "Too much is never enough." MTV, 1983
  • Ha... (Score:3, Funny)

    by Sensible Clod (771142) <dc-7@cWELTYharter.net minus author> on Monday April 25, 2005 @03:53PM (#12339929) Homepage
    500 terabytes -- would take about 250 years to download using a typical 512 kilobit per second household broadband connection

    Well, I've got a 3 megabit connection! It'd only take...uh...well, 42 years or so...but I'd upgrade to that 1 gigabit connection they have in Asia before it finished...
  • What was not revealed in the article, was that the majority of the data was composed of pictures of Goatse and TubGirl in ultra-high resolution..
  • At that rate it would still take a little over 3 years to give every family in the USA one copy of SimplyMEIPS from that single pipe.

    Now, with a dozen pipes like that the task could be done in a month....
  • by Quixote (154172) * on Monday April 25, 2005 @03:57PM (#12339965) Homepage Journal
    What is with these non-standard terms like "Terabytes" and "Megabytes"? Please re-state the bandwidth and the amount of data tranferred in LoCs (Libraries of Congress) and KLoCs (Kilo Libraries of Congress) so that the rest of the world can understand the magnitude of this achievement.
    • Should they transfer 1024 Terabytes and store it all in one chunk, they will have a PetaFile.
    • Re:Standard terms (Score:2, Interesting)

      by The_Wilschon (782534)
      Actually, that raises the interesting point of whether they transferred 500 terabytes (500 x 10^12 bytes / 500000000000000 bytes) or 500 tebibytes (500 x 2^40 bytes / 549755813888000 bytes)? In the kibibyte/kilobyte range which prefix you use doesn't make much difference (~2.4 % difference), but at tebi/terabyte range it kinda does (~9.5 % difference).
  • I have a broadband connection. My laptop is always online. But what do I do? Check email, check /., check email again, check /., check /. THATS IT! (and ofcourse a bit of surfing here and there). The current bandwidth is kind of sufficient. I wish I had more when I try to download movies or files, but then I can live with it.

    Imaging 2007, *AA has made it almost impossible to download any content. So I'm sitting on 600 MB/sec of BW and checking /. and reading emails.

  • It's good that they're speeding this thing up. One day, all homes will have this kind of broadband connection coming in. This will be a necessity because all television programming, telecommunications, and other functions will take place over this network. When you rent a movie, you won't have to wait for a DVD to come in the mail. It will instantly be there and ready for you to watch. When you download pr0n, you won't have to wait for it to get there, losing your current state of passion. This is going to
  • The total amount of data transmitted during this challenge -- 500 terabytes -- would take about 250 years to download using a typical 512 kilobit per second household broadband connection

    Thats why I d/l at 6 Mbits/ second.
    • Thats why I d/l at 6 Mbits/ second.

      So that would only take you... what, only 20.83 years or so? Yeah, that's WAY better.

      I can't understand the measurements they're using... can someone state this in terms of Libraries of Congress per second?
      m-
      • Well according to wikipedia its estimated, that the library of congress (in text) is about 15-17 terrabytes. So that ranges from about 29 - 33 Libraries of Congress.
  • We all know that scientist have HDTV and the good pron!
  • by rnxrx (813533) on Monday April 25, 2005 @04:13PM (#12340119)
    600 MB/sec = 4.8 Gb/sec.

    OK... they lit up the equivalent of two OC48's worth of bandwidth. That's half of an OC192 or a 10G Ethernet. There have been long haul OC192's for a number of years now. If I hook up a hardware-based traffic generator and run at 100% over an OC192 for a few weeks will I get a slashdot article, too?

    • If I hook up a hardware-based traffic generator and run at 100% over an OC192 for a few weeks will I get a slashdot article, too?

      If you do so on a worldwide scale and use that infrastructure to solve the fundamental questions of physics, yes. It's not supposed to be that technologically amazing, it's just a progress report and a proof of concept.

    • by stefanb (21140) * on Monday April 25, 2005 @04:40PM (#12340426) Homepage
      OK... they lit up the equivalent of two OC48's worth of bandwidth. That's half of an OC192 or a 10G Ethernet. There have been long haul OC192's for a number of years now. If I hook up a hardware-based traffic generator and run at 100% over an OC192 for a few weeks will I get a slashdot article, too?

      The accomplishment is not in the data rate, it's in the ability of the participating organisation's to get a stable network going. One that is close enough to the one that the real scientists will be using in a couple of years.

      Consider that there's a large number of institutes, universities, etc. that all have their own IT departments, plus all the physicists that have to be involved because it's their grant money funding all this. It's thousands of people coordinating. And I would be surprised if they hadn't set up different service classes, priorisation schemes and what not.

      Setting up a trivial network between a couple of sites that are all under your control is close to trivial: you just need to talk to you telco and buy the lines, and hook up the routers. But establishing a working network between these many institutions takes a lot more.

    • They sustained it. Via several hops across the atlantic, over a week.

      This is important because the experiment will create more data than can be stored there, so it has to be dumped "into the grid". If the datarate couldnt be sustained, data would be lost.
    • If I hook up a hardware-based traffic generator and run at 100% over an OC192 for a few weeks will I get a slashdot article, too?

      Probably yes, if you route the traffic to a major e-commerce site.

      5 Gbps sustained is indeed not that much traffic, and the underground grid computing community probably has gone beyond that by now.
  • MPAA & RIAA (Score:5, Funny)

    by killtherat (177924) on Monday April 25, 2005 @04:15PM (#12340142)
    On a related note, CERN is now being sued by the MPAA & RIAA. A spokesmen was commented, saying, "Obviously with 500 terabytes of data being transmitted on the internet, at least some of it had to be copyrighted materials represented by the RIAA and the MPAA. As we know, the internet and communication grids serve no real purpose other then to pirate movies and music."
    The lawsuit is expected to destroy CERN and any sort of decent networking research anybody was even thinking about doing for the next 50 year.
    • A small canister with the CERN logo was found in the home of the RIAA/MPAA. According to security officials, "We cannot find this canister as it is hidden, and the culprit utilized one of our wireless cameras. The interesting thing about this transparant container with a count-down timer is that it has a floating ball of what looks like silver in the middle of it."
  • by millennial (830897) on Monday April 25, 2005 @04:23PM (#12340207) Journal
    Time required by my home internet connection, 3mbps, to transfer this data: 41.6666666... years. Rounded to one sig fig, since 500TB is: 42 years. It really IS the answer to life, the universe, and everything.
  • Lofar project (Score:4, Interesting)

    by photonic (584757) on Monday April 25, 2005 @04:31PM (#12340302)
    ... the most data-intensive physics instrument on the planet. Today eight major computing centers successfully completed a challenge to sustain a continuous data flow of 600 megabytes per second ...
    I don't know how fair the comparison is, but I think the Lofar [lofar.org] project will be a heavy contestant for the claim of the experiment with the highest data rate. It is basically a array of some 10000 radio antennas, spread over the northern part of the Netherlands and Germany. It will be operated as one huge phased array.

    The data rate [lofar.org] might even be bigger than at Cern: 20 terrabit/sec straight after the A/D converters and still a mighty 0.4 terrabit/sec after the initial data reduction (DSPs + FPGAs). All the remaining data will be transfered over a dedicated fiber network to a central computer. To reduce all this data they need a big fat supercomputer, this will be a IBM Blue Gene [ibm.com] with serial number 2, to be handed over tomorrow [zdnet.com]. For the moment it will be the fastest computer in Europe and ranking somewhere in the top 10 of the world.
  • by Sri Ramkrishna (1856) <[sriram.ramkrishna] [at] [gmail.com]> on Monday April 25, 2005 @04:42PM (#12340465)
    Seth Lord and RIAA Chief Mitch Bainwol, felt a sudden disturbance in the force. It was a like a thousand music producers and label execs suddenly cried out in grief and dispair.

    sri
  • by Eyeball97 (816684) on Monday April 25, 2005 @04:47PM (#12340518)
    It would only take 247.73274987316083206494165398275 years. I wish these articles would check their facts. 247.73274987316083206494165398275 I could live with. I think at 250 I'd start to get impatient...
    • 247.73274987316083206494165398275 I could live with.

      Yeah, and if you factor leap years into your calculation, the number is 247.56318604710117372677263163232 ;)
  • one must wonder what that 500TB consisted of that was transferred. was it the same file over and over again? was it just one gigantic file? was it randomly generated bits? or was it... bittorrent?
  • by brxndxn (461473)
    PORN

    eom

    (can't believe I haven't seen one of those with a FUNNY yet)
  • I wonder how much of that 500TB wasn't pr0n.
  • Thanks to Speakeasy, it'll only take 20 years to download this at my house! Suckers!
  • Don't you think.
  • how many libraries of congress per hour is that?

Their idea of an offer you can't refuse is an offer... and you'd better not refuse.

Working...