Slashdot Mirror


Researchers Transmit Optical Data at 16.4 Tbps 2550km

Stony Stevenson writes "The goal of 100 Gbps Ethernet transmission is closer to reality with the announcement Wednesday that Alcatel-Lucent researchers have recorded an optical transmission record along with three photonic integrated circuits. Carried out by researchers in Bell Labs in Villarceaux, France, the successful transmission of 16.4 Tbps of optical data over 2,550 km was assisted by Alcatel's Thales' III-V Lab and Kylia, an optical solution company. The researchers utilized 164 wavelength-division multiplexed channels modulated at 100-Gbps in the effort."

126 comments

  1. Translation please? by davidwr · · Score: 5, Funny

    What's that in Library-of-Congresses per fortnight?

    --
    Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
    1. Re:Translation please? by CRCulver · · Score: 4, Funny

      Is that a laden European or African library of congress?

    2. Re:Translation please? by pizzutz · · Score: 2, Informative

      I calculate roughly 248,000 Library of Congresses per fortnight.

      Curse my geeky genes for making me calculate that when you asked.

      --
      GE/CS/IT d- s: a- C++++$ UL+++ P-- L++++ E W+++$ N+ o? K- w---() !O M- V- PS+ PE(++) Y+ PGP+++(+) t+++ !5 X++> R- t
    3. Re:Translation please? by TheScorpion420 · · Score: 1, Insightful

      Or how many 747's full of DVD's?

      --
      If you pay your taxes you support terrorism!
    4. Re:Translation please? by suso · · Score: 2, Funny

      HD DVD, Bluray or regular?

    5. Re:Translation please? by thatskinnyguy · · Score: 1

      Either way, I think my file server needs more space.

      --
      The game.
    6. Re:Translation please? by Eddi3 · · Score: 5, Informative

      (14 * 24 * 60 * 60) / (20 / 2.2) = 123,984 LoCs/fortnight

      (total seconds per fortnight)
      14 days per fortnight
      24 hours per day
      60 minutes per hour
      60 seconds per minute

      all over

      (seconds per Library of Congress transferred)
      20 terabytes per second (one LoC/second)
      2.05 terabytes per second (16.4 terabits per second

    7. Re:Translation please? by aplusjimages · · Score: 1

      Let's just say it's a lot of pr0n.

      --
      Can I bum a sig?
    8. Re:Translation please? by Anonymous Coward · · Score: 0
    9. Re:Translation please? by michrech · · Score: 1
      I prefer HHDD-DDVVDD, thank you.

      HD DVD, Bluray or regular?
      --
      bork bork bork!
    10. Re:Translation please? by pizzutz · · Score: 1

      I estimated a LoC as 10 TB (first number I saw after quick googling), which explains why my number is twice as big. At least my math was right :)

      --
      GE/CS/IT d- s: a- C++++$ UL+++ P-- L++++ E W+++$ N+ o? K- w---() !O M- V- PS+ PE(++) Y+ PGP+++(+) t+++ !5 X++> R- t
    11. Re:Translation please? by a_claudiu · · Score: 1

      But can you calculate the flux of in information in the cable in LoC/(square furlong * lapse year)? Once you go metric you never go back.

    12. Re:Translation please? by Eddi3 · · Score: 1

      Well, I'm not exactly sure what you're asking, but here we go:

      (((650 / (0.0014 * .12 * .12)) * 8.54) / 1024) / 1024 = 262.591574 Petabytes

      650 cubic meters - rough volume of 747 [1] .0014 meters - rough thickness of DVD .12 meters - diameter of DVD
      8.54 GB - dual layer DVD capacity
      1024 GB in 1 TB
      1024 TB in 1 PB

      So, a 747 can carry about 250 Petabytes of data in Dual Layer DVDs at a time. Then just divide that by the time it takes to fly it wherever you want it.

      [1]: All volume values for the 747 were found at http://www.zap16.com/civ%20fact/civ%20B747-300.htm (model 747-300)

    13. Re:Translation please? by networkBoy · · Score: 2, Informative

      Data per boeing 747 (LCF version)

      DVDR = 159238213.7 GB/747LCF
      HDVD = 677609420 GB/747LCF
      BDVD = 847011775 GB/747LCF

      -nB

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    14. Re:Translation please? by u-235-sentinel · · Score: 4, Funny

      What's that in Library-of-Congresses per fortnight?

      Well... if you are Concast they will give you those numbers in terms of photos or mp3's or emails downloaded in a month.

      Personally I like to know in terms of how many 8 track tapes I can download a month. ;-)

      --
      Has Comcast disconnected your Internet account? Same here. You can read about it at http://comcastissue.blogspot.com
    15. Re:Translation please? by Anonymous Coward · · Score: 0

      Informative? That can't be right, that's a 17714 libraries per 24 hours, or about 1 library per 5 seconds. Those Library Of Congresses would have to be about 10 TB a piece. Yet wikipedia tells us it is estimated at 17 to 20 TB of information, so it's closer to 125000 Library of Congresses per fortnight.

    16. Re:Translation please? by bettega · · Score: 1

      At last now I know what a fortnight is. It is a quinzena :)

    17. Re:Translation please? by Anonymous Coward · · Score: 0

      All of them.

    18. Re:Translation please? by Dagger2 · · Score: 1

      Unfortunately, according to that page, the max payload weight is 66,300kg, and DVDs weigh about 17 grams, meaning you're limited to 3.9 million DVDs = 31.763 PB.

      Unless you plan on taxiing all the way there.

    19. Re:Translation please? by VE3MTM · · Score: 1

      No, the real question is: is it faster than a station wagon full of Blu-Ray discs on the highway?

      --
      09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0 Whoops, silly middle mouse button...
    20. Re:Translation please? by Anonymous Coward · · Score: 0

      They'll give you those figures alright, but then they'll delete your routings after an unpublished number of torrents per hour are accessed.

      After not having my connection drop for several months, I've been having to power cycle my cable modem just because my roommate's been running a torrent client the last week or so... even though the status lights say everything's peachy.

      And I'm paying for the premium cable modem service (the most expensive residential plan.)

      I hate their legal monopoly and "we'll do what we please" terms of service.

    21. Re:Translation please? by Anonymous Coward · · Score: 0

      You sure you ain't a ter'ist? Thought I heard you say "laden".

    22. Re:Translation please? by thechao · · Score: 4, Funny

      I saw "LoC" and thought "Lines of Code"; my second thought was that you have some *mad* job security.

    23. Re:Translation please? by JrnyFan · · Score: 1

      It depends on if they grip it by the husks or hold it under the dorsal feather I hear.

      --
      If the prevalent philosophy is that life is a figment of my imagination, why didn't Martha Stewart get the chair?
    24. Re:Translation please? by fbjon · · Score: 2, Interesting
      Your calculations are a bit off. The LCF, like the Beluga and similar, is meant to transport aircraft parts which are large, not heavy. Additionally, the bulky airframe means it can actually lift less weight than a regular cargo carrier, and maximum takeoff weight is the limitation for bandwidth, not volume. Besides, the LCF is not for sale to customers.


      Redoing for the 747-400ERF:

      • Assume each disc weighs 16g, like a CD.
      • This gives us a box with a volume of 1,38 m^3 that contains 80000 discs, weighing 1280 kg, let's say 1300 kg including the box.
      • With a maximum payload of 112760 kg, that means 86,7 boxes, giving a fuel range of 9782 km. Note that if you want to go the full 14212 km, you'll have to throw off some weight to load more fuel.
      • This means about 6 936 000 discs giving the following jumbo packet sizes:
      CD = 4,5 petabytes
      DVD(DL) = 51 PB
      HD-DVD = 184,8 PB
      BD = 308,3 PB
      • Not including landing and takeoff times, the aircraft will travel 9782 km in about 10 hours, but let's add in 30 minutes for loading and unloading, giving the following practical bandwidths:
      CD = 124.8 GBps
      DVD(DL) = 1.38 TBps
      HD-DVD = 5 TBps
      BD = 8.35 TBps


      Now, according to wikipedia the Airbus A380F has a maximum range of 3800 km less, but has a maximum payload of about 37240 kg more, and would thus be better for bandwidth over normal distances as opposed to extreme long haul transmission. The calculations for this are left as an excercise to the reader.

      --
      True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
    25. Re:Translation please? by kesuki · · Score: 1

      I'm assuming you're using dual-layer media, but technically, bd-rom and such support Dual sided, dual layer media, as dvd also supports. of course finding writable dual layer dual sided media is different, but assuming you used standard manufacturing of pressed blue ray drives, you could use dual sided dual layer media. that effectively doubles both your figures for bd-rom and dvd-rom (unless you considered the dual sided capabilities i didn't check your dvd numbers)

    26. Re:Translation please? by tenco · · Score: 1

      Silverlight or HTML?

    27. Re:Translation please? by slashqwerty · · Score: 1

      How about channels of high definition video? High def television broadcasts are 20 Mbps. So this technology can simultaneously transfer 820,000 high def channels over a single optical fiber.

  2. So, this speed, on a scale of 1-10... by r_jensen11 · · Score: 4, Funny

    Would this qualify as 11?

    1. Re:So, this speed, on a scale of 1-10... by suggsjc · · Score: 1

      For now...wait a couple of years and it will be a measly 5-6. Not sure what we are going to do with all of that bandwidth, but I'm sure that we'll come up with something interesting and possibly even useful.

      --
      When I have a kid, I want to put him in one of those strollers for twins and then run around the mall looking frantic.
  3. CPU speeds? by Anonymous Coward · · Score: 1, Interesting

    Surely there must be some incredibly processing power behind transmission speeds like this? Anyone one have any idea?

  4. ObWalken by Saint+Aardvark · · Score: 3, Funny
    1. Re:ObWalken by Hybridan · · Score: 1

      Be honest, how many of you followed the link above, but the first thing you looked at was the picture with the two women and Dwayne Johnson. (Ok, now be honest, how many of you followed the link after I mentioned Dwayne Johnson or the two women?)

  5. Don't get too excited. by geminidomino · · Score: 5, Funny

    That's just BURST throughput. Depending on factors like time of day, how many other users there are, and environmental conditions, throughput may drop as low as 33kbps. And we do NOT filter bittorrent.

    Just check your TOS agreement. It's all right there.

    1. Re:Don't get too excited. by holyspidoo · · Score: 5, Funny
      Now offering the fastest 16.4 Tbps service* available anywhere

      *1 Gig upload/download monthly limits apply

    2. Re:Don't get too excited. by BoredAtWorkWhatElse · · Score: 1
      I know you're joking but my ISP is now offering a 50Mbps with a 50GB limit download/upload combined. 1.50$ per additional GB, no limit.

      Remind me again who needs a connection that can only work at 100% capacity for a little more than 2 hours per month ? Only to be charged 50c/s afterward (no limit !!!)

    3. Re:Don't get too excited. by gravis777 · · Score: 1

      Pitty the net is so slow. 16 Tbps connection from the phone company to my house, yet download speeds still rarely top 600k a second. Difference between now and before is I can now download 1,000 files in 30 minutes instead of just one file in 30 minutes. Sigh. Of course, with Bittorrent and dedicated Usenet servers, I can get much faster downloads. Hmph, maybe iTunes will start offering downloads at 320kbps on their songs and full 1080p on their movies.

  6. On Neutrality by Anonymous Coward · · Score: 2, Insightful

    In other news: American telcos wonder how French providers are able to afford research and development without additional funding from a tiered billing billing scheme that is needed to advance the science in the United States.

    1. Re:On Neutrality by mrvan · · Score: 1

      Never applied for EU research funding, did you? :-)

  7. Current cables? by headkase · · Score: 1

    With the ever-predicted bandwidth crunch always just around the corner, can existing cables be reused just by replacing the signaling equipment at substations with this? If we don't have to lay all new cables - just upgrade the nodes - then upgrading to these bandwidth capacities on our current networks would be a cinch.

    --
    Shh.
    1. Re:Current cables? by afidel · · Score: 2, Informative

      The problem has never been the glass! There is absolute craploads of dark fiber just about everywhere. Last time I saw stats it was something like less than 1/3rd of installed fiber was lit up. It's the uber expensive routing equipment needed to keep up with the flood of data that's the expensive part.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    2. Re:Current cables? by Anonymous Coward · · Score: 0

      The dispersion of the glass needs better control and the gain flatness across the band in use has to be up to scratch as well. Not all erbium amlifiers are good enough for an upgrade like that.

    3. Re:Current cables? by duncan99 · · Score: 1

      That, and the last mile (or even the last 25 feet) to YOUR house.

      (I've always seen that quoted as "Five boxes preserve our freedoms: soap, ballot, witness, jury, and cartridge" But yeah, that.)

    4. Re:Current cables? by kylehase · · Score: 1

      This is true. While these new advances in high capacity trunk lines are great it's not noticeable for end users if their connections to the trunks the bottleneck. There needs to be a bigger push for fiber to the home.

      --
      You want fun, go home and buy a monkey!
  8. obligatory Homer J. Simpson quote by ionix5891 · · Score: 0

    "i have invented a way of downloading porn a million times faster!"

  9. maybe its just me by the_mind_ · · Score: 5, Funny

    "164 wavelength-division multiplexed channels modulated at..."

    how very Star Trek of them.

    --
    You feel sleepy. Close your eyes. The opinions stated above are yours. You cannot imagine why you ever felt otherwise.
    1. Re:maybe its just me by pizzutz · · Score: 5, Funny

      "164 wavelength-division multiplexed channels modulated at..."
      how very Star Trek of them.

      I'm sorry Captain, but we canno' reach these speeds with time-division multiplexing. the phase coils canno' handle it!
      --
      GE/CS/IT d- s: a- C++++$ UL+++ P-- L++++ E W+++$ N+ o? K- w---() !O M- V- PS+ PE(++) Y+ PGP+++(+) t+++ !5 X++> R- t
    2. Re:maybe its just me by funaho · · Score: 1

      A soon as they can get the multimodal reflection sorting working we can tap into the Borg collective. They have all the good warez, pr0n, music and movies from thousands of civilizations. :)

    3. Re:maybe its just me by Teiresias_UK · · Score: 2, Interesting

      To be honest I don't find this *that* amazing.

      I worked at the victim-of-the-telecoms-bubble that was Marconi 2000-02 and there was a bit of kit, the snappily titled UPLx, that could deal with 160 10Gbps channels down a pair of fibres, unregenerated over about 1000km - using soliton wave shaping and some sodding great Ramen pump laser to get there (nothing to do with noodles before you ask). It was demoed in the labs reliably, and I believe sold in to Telstra Australia

      In 5 years, they've added 4 Gbps ... wow.

    4. Re:maybe its just me by MadUndergrad · · Score: 1

      You're off by an order of magnitude. They've added about 15Tbps.

    5. Re:maybe its just me by fbjon · · Score: 1

      4 Gbps as well as 1500 km, mind you.

      --
      True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
  10. Doesn't matter by Lumpy · · Score: 0, Troll


    No broadband ISP will upgrade even if 100Gbps was cheap, they are happy with keeping everyone in their low speed high latency world. But they will start using more buzzwords and charging more!

    Now with rocketsmell!

    --
    Do not look at laser with remaining good eye.
    1. Re:Doesn't matter by dave420 · · Score: 2, Insightful

      Surely if an ISP adopted this, they'd have people signing up left right and centre. Wouldn't it be awfully attractive to their target audience?

    2. Re:Doesn't matter by darthflo · · Score: 1

      Unfortunately both target audiences (that's 1. Supercomputing/clustering big/experimental stuff and 2. being a Tier 1 backbone provider) for this kind of technology don't need their ISP to provide it. Your average SoHo couldn't even deliberately use a single Gigabit link; even well-connected datacentres don't (yet) need 100 Gbps outside connectivity.

  11. 16.4 Tbps of optical data? by Gabest · · Score: 2, Insightful

    Tbps speed, and over 100 Gbps. Something is wrong here.

    1. Re:16.4 Tbps of optical data? by spectrokid · · Score: 5, Informative

      They had 164 lasers with different colours sending 100 Gbps EACH over the same fiber, splitting the colours apart again at the other end with what probably is a little more advanced than a prism.

      --

      10 ?"Hello World" life was simple then

    2. Re:16.4 Tbps of optical data? by knight24k · · Score: 0

      Optical data...hmm...did they just download all the pr0n on the internet at once?

      Can I get a copy of that?

    3. Re:16.4 Tbps of optical data? by EricR86 · · Score: 2, Informative

      I'd wadger they're using devices like a Diffraction Grating or a Fabry-Perot Etalon

      Only a little more complicated than a prism :)

    4. Re:16.4 Tbps of optical data? by HTH+NE1 · · Score: 1

      Not all the pr0n, just the really fast pr0n.

      Measuring data quantity by its speed over a distance in kilometers? What's next, measuring its speed by its acceleration over a volume in Liters?

      --
      Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
    5. Re:16.4 Tbps of optical data? by colinmcnamara · · Score: 5, Informative

      Its called dense wave division multiplexing, or DWDM. You take independent links (in this case 100Gig links), and transmit each of them on a slightly different wavelength of light called a Lambda. Since optic is looking for a specific wavelength, you can now run many "virtual links" per physical fibre. This is the standard technology for most Telcos. The innovation here is that they are doing this with 100Gig transceivers, and they have chipsets fast enough to combine the different lambda's back together into on high speed link. And yes, you can now let the Lambda Lambda Lambda jokes fly

      --
      Colin McNamara - CCIE #18233 "The difficult we do immediately, the impossible just takes a little longer"
    6. Re:16.4 Tbps of optical data? by Adeptus_Luminati · · Score: 1

      100 Gbits/sec * 164 laser colours sent over a single pair of fibers = 16,400 Gbits/sec total throughput.
      16,400 Gbits/sec = 16.4 Terabits/sec
      16.4 Terabits/sec / 8 bits = 2.05 TeraBytes /sec

      If you think about it, this is like shipping 2x 1Terabytes hard drives accross the Atlantic in 1 second.

      Now if you want the SneakerNet comparison... or perhaps JetNet in this case....
      Say you could fit 10,000 x 1TeraByte hard drives in a 747 air plane, and it takes you 6 hours to cross the Atlantic.
      6 hours= 21600 seconds
      10,000 hard drives / 21600 seconds = 2.16 Terabyte hard drives per second

      So in short, they have been able to finally catch up with the Truckloads of DVDs or in this case JetLoads of Hard drives comparisons.Whoop-de-do! They would certainly fail the 50GB/disc Blue-Ray over Jet analysis though as you can fit more of those than hard drives.

      On the plus side, you don't waste fuel and no jet lag, so I'm all for it! hehe

      --
      No trees were killed in the making of this post; however, many trillions of electrons were horribly inconvenienced.
    7. Re:16.4 Tbps of optical data? by Anonymous Coward · · Score: 0

      well it is a bit better than that as they have the 100gb channels spaced at 10Ghz apart.

  12. error checks? by sjs132 · · Score: 1, Interesting

    Slim article... How long would it take to error check that much data?

    on Another note... What did they do with all that Pr0n once it got to the other end?

    --
    --- Relax, that mass muderer is just trying to reduce our carbon footprint, one fetus at a time...
  13. yeah, this would fill my hard drives in .03 second by VisiX · · Score: 1

    But what is the latency involved in a system like this? Right now I get about 25000kbps and that is plenty for me, but what I really need is reduced latency for real time simulations.

  14. Doesn't matter... by ScaryMonkey · · Score: 5, Funny

    No matter how much speed they create, they will still be subject to the Law of Diminishing Porn Returns, which states:

    For download rate n, my demand for new porn will require me to download at a rate of n+1.

  15. Sending "optical data" by Splab · · Score: 1

    Where I the only one thinking a truck filled with DVD's when the headline said optical data?

    1. Re:Sending "optical data" by thwack328 · · Score: 4, Funny

      Senator Stevens? Is that you?

  16. Make it Short and Fast and Snappy by Doc+Ruby · · Score: 0, Flamebait

    Why do we need 100Gbps-e to arrive across 2550Km before it's "reality"? All I want is 100Gbps-e over maybe 100m, or even 10-30m to start. Since there's no other local interconnect faster than 10Gbps-e, I'd even settle for 100Gbps-e across 1m, for interconnecting in my rack without changing my software that all depends on ethernet between hosts.

    What's surprising is that I can get 10Gbps-e for something like $50, but nothing faster for any higher price (except multiple 10G-e cards). AFAICT, there's nothing in the engineering pipeline, like the usual faster interconnects for supercomputers that can be ported to PC buses the way SCSI and optical ethernet (etc) were. Everyone's waiting on 100Gbps-e.

    I know that the parties funding the 100Gbps-e research are telcos like Alcatel, which have WAN requirements for thousand-Km hops, and governments encouraging the industry at the top end. But broadband projects around the world are planning to roll out 1Gbps and even 15Gbps fiber to the home, which would completely saturate a 100Gbps trunk after much shorter neighborhood distance to a hub than 2500Km, even in the most rural areas. But what about all the LAN vendors, which have a real market for 100Gbps-e with only a few dozen or hundred meters required between nodes?

    Can't we just declare success and start buying some fast ethernet without having to satisfy the telcos' 10 year plans, too?

    --

    --
    make install -not war

    1. Re:Make it Short and Fast and Snappy by Anonymous Coward · · Score: 1, Informative

      Consider a standard computer utilizing two channels of ddr2-800. This only has a bandwidth of of 102.4 gbps. You would need to attach this connection directly to your ram in this case. Now computers are starting to utilize ddr3 which can theoretically provide double that in the same configuration. But at those speeds, amount of available ram would be limiting the actual throughput. 100gbps utilizing 128GB of ram would be emptied in 10.24s (1024gb/100gbps). Having to have a storage array capable of sending out 100gpbs continuously would require at least 125 drives at 800mbps(100MB/s is a good number) constantly. This is the reason 100gpbs isn't being considered for lan use. It just isn't feasible at this point.

    2. Re:Make it Short and Fast and Snappy by D4rk+Fx · · Score: 1

      Where can you get 10G Ethernet cards for $50? I'd really like to know and I'm sure there would be a lot of interested people...

    3. Re:Make it Short and Fast and Snappy by Gaima · · Score: 2, Insightful

      This is the reason 100gpbs isn't being considered for lan use. It just isn't feasible at this point.

      Stick a thousand machines on each end, and you'll understand why 100Gbps is needed.

    4. Re:Make it Short and Fast and Snappy by Anonymous Coward · · Score: 0

      > Since there's no other local interconnect faster than 10Gbps-e,

      Yes there is, Infiniband 4x DDR at 20Gbps, or 12x at 30 (60 for DDR) Gbps.

    5. Re:Make it Short and Fast and Snappy by afidel · · Score: 2, Informative

      If you really need greater than 10Gbps then go with Infiniband as you can get 12x HCA's that will do 24GBps (48Gbps full duplex). But if you're paying $50 for 10Gbps ethernet you're not getting offloading and your CPU's are probably swamped of your TCP/IP stack is the problem. I would suggest getting a pair of offloading 10Gbps cards and seeing if you don't see a huge improvement.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    6. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 2, Interesting

      I'm running Linux on a Playstation3 with SPU video drivers in its Cell uP that can run at over 150GFLOPS. Since the PS3 has only 512MB RAM, it needs to be fed by the LAN and just buffer the LAN in its RAM. Even if SATA drives are delivering only 1.2Gbps, there's no reason I can't have multiple parallel drives on independent servers (if a single server's IO isn't fast enough for multiple SATA at full bore) on my SAN delivering multiple streams through my switch all to my Playstation. Now, the PS3's 1Gb-e is as hardwired to it as is its 512MB RAM, but the point is that there are already machines that can use that bandwidth. The total bandwidth doesn't have to reach 100Gbps, but only exceed 10Gbps, to require faster than 10G-e, which only 8-10 SATA drives in parallel could do today.

      So the bottleneck is 10G-e. There are already supercomputer clusters using multiple parallel Cells, so I'm disappointed that they're not already widening the pipe.

      --

      --
      make install -not war

    7. Re:Make it Short and Fast and Snappy by leomaro · · Score: 5, Interesting

      There is another problem, and is actually the bottleneck of transmitting packets at high rates.

      It doesn't really matters (yet, and considering Ethernet technology) if the BW of the fiber is a zillion Petabits/sec.
      The problem is now at 1Gbps and 10Gbps in Ethernet technology, and is because the processor overloads with the amount of hardware interrupts. The processors that are general purpose have to waste too many clock cycles processing that much interrupts, the processors nowadays are superscalar [ http://en.wikipedia.org/wiki/Superscalar ]and every time the processor have to change the context (to attend an interrupt) has to do lots of things like unloading the registers, saving the context, loading the registers of the new process, and has to drop something out of the pipeline [ http://en.wikipedia.org/wiki/Pipeline_(computing) ] loosing performance.

      Ethernet tech has a huge latency [ http://en.wikipedia.org/wiki/Latency_(engineering) ] and a stack that makes processing not so easy (if you look at te code of a linux network device driver it handles pretty much everything including writing the mac address that is only copied when the driver initialize).

      That is why there are some relative new things (NAPI in Linux) that try to make lessen the overload, there are new network devices that handle layer 2 and 3 (or at least parts of those, for example, is used to be handled the checksum algorithm) to avoid doing it in the processor. There are some white papers (one from intel, another from NetXen, I'm sorry I don't have the links now) that explain the problem and some approach to a possible solution.

      Yes, I know, there is something I have not said, and is that the main switches or routers have to deal with that and have hardware specially designed to do heavy network packet processing, and that is the point, the network cards will have to do that (and are already starting to), neither is an easy job for hardware designers, nor for the market, is easier and cheaper to have a machine that you can change the behaviour only changing the firmware or changing settings from a program (routers have an operating system, and lots of those are a general purpose microprocessor with a linux kernel and a web server to configure it, for example home routers).

      There is much to say yet in this field.

    8. Re:Make it Short and Fast and Snappy by Moskit · · Score: 2, Informative

      "Everyone" is waiting for either 40Gbit/s or 100Gbit/s Ethernet.
      The first one is what server-people push (they claim they do not need more, that's why 40Gbit/s was put into Ethernet standard),
      while network people want full 100 Gbit/s.

      > But what about all the LAN vendors, which have a real market for 100Gbps

      They don't.
      There seems to be market either for 40Gbit/s in LAN/local connections or for 100Gbit/s for core/long haul. At least judging but what happened with high-speed ethernet standard.

    9. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      I had found with Froogle a Sun 10G-e card (probably refurbished) for $35, but it's not available now on searching again. It does seem that 10G-e cards do cost something around $500 at least, and plenty up around $900+.

      But the point is that I can't buy anything faster for even 10x or 100x, except multiple cards. And maybe some really exotic interconnect that's not ethernet, so apps have to be recoded to use it.

      --

      --
      make install -not war

    10. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1
      True, but the rest of the sentence said:

      Since there's no other local interconnect faster than 10Gbps-e, I'd even settle for 100Gbps-e across 1m, for interconnecting in my rack without changing my software that all depends on ethernet between hosts.


      Infiniband isn't ethernet.
      --

      --
      make install -not war

    11. Re:Make it Short and Fast and Snappy by Anonymous Coward · · Score: 0

      Yes there's certainly a demand, but going over 10 Gbps is actually quite a difficult thing to do. The circuity required to make a laser diode blink properly at 10 Gbps is pretty involved. Plus, as of present, said circuity must be fabricated on a separate semiconductor substrate than the diode. Then there are issues of heat dissipation and EMI... I've never messed with wavelength multiplexing, but I'm sure that is wildly difficult and expensive too.

      But yes, the demand is certainly there. And I can promises you that there are hoards of teams of photonics researchers slaving in cleanrooms and over bit error rate testers in the hopes of filling that demand. Its probably a few years down the pipe to any breakthrough though...

    12. Re:Make it Short and Fast and Snappy by Detritus · · Score: 2, Informative

      Interrupt loads can be greatly reduced by switching to a polling-driven architecture. See FreeBSD.

      --
      Mea navis aericumbens anguillis abundat
    13. Re:Make it Short and Fast and Snappy by afidel · · Score: 1

      You're apps don't talk to Ethernet, they talk to your TCP/IP stack! There is an IP over Infiniband standard which is supported by most HCA's and it's even fully offloaded to the HCA processor. IPoIB isn't as efficient or low latency as native Infiniband protocols but at least you don't have to recode your apps =)

      And the reason there isn't anything faster is that it's such an incredibly niche market, the number of sites that need greater than 1GB/s on a single link are very, very small. Heck the storage for even big servers is generally only 4Gbps per link as 10Gb FC is still astoundingly expensive (Cisco 10Gb FC blade for instance is $60K for 4 ports) so it's cheaper to just put a couple 4Gb cards in and get additional redundancy for free.

      P.S.
      This is post #5,000 for me =)

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    14. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      Well, it turned out that 10Gbps interfaces actually cost more than $50, probably more like $500 per port. But they don't require recoding apps to use the Infiniband fabric.

      The point is really that there's a huge jump from a $15 1Gb-e (or $500 10Gb-e), if that jump can be made at all, while there's no 100Gb-e at any price. And instead of rolling out 100Gb-e that works for LANs, the industry is evidently waiting until it's good for continent-spanning WANs.

      --

      --
      make install -not war

    15. Re:Make it Short and Fast and Snappy by Torg · · Score: 1

      In my experience I have great demand not for 100Gbps Ethernet, but Fibre Channel. It is a common problem is replicating data between data centers. The high end approach to this is to get two DWDM switches and dark fiber (if you can afford it).

      The main problem with transporting large data segments over Ethernet is quite simple. Data is read off disk (typically) at 8KB chunks, then converted to 1.5Kb Ethernet packets. This "fragmenting" and additional overhead is what causes allot of the resultant lag.

    16. Re:Make it Short and Fast and Snappy by afidel · · Score: 1

      You keep talking about recoding apps, YOU DON'T NEED TO. There is a standard for IP over Infiniband. It's not as performant as native Infiniband protocols but it DOES exist and is in use. Besides without a dedicated specialized processor and gobs of buffer memory your typical server isn't going to be able to keep up with 10Gbps Ethernet let alone anything faster.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    17. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      How can it be a few years until "any breakthrough" in >10Gbps when we're discussing successful tests of 100Gbps across 2550Km already? Sounds like the breakthroughs have already arrived, but telcos are aiming for a totally different market than LANs. If it's working even most of the time at 2550Km, then it should be pretty close to commercial availability at 10m or 100m.

      --

      --
      make install -not war

    18. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      I don't see why a fast ethernet card can't have its own DSPs and/or FPGAs for decoding packets, then sending them across PCI-e to XDR memory. For $500 or $1000, 100Gbps HW should be able to deliver.

      --

      --
      make install -not war

    19. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      Well, "100"Gbps is an arbitrary number. So long as it's significantly bigger than 10Gbps. 40Gbps would be good, too, but it's not available either.

      The advantage of ethernet is that apps are already coded to use it. Including all the network mgmt apps.

      --

      --
      make install -not war

    20. Re:Make it Short and Fast and Snappy by leomaro · · Score: 3, Insightful

      >Interrupt loads can be greatly reduced by switching to a polling-driven architecture. See FreeBSD.

      New API (NAPI) takes a mixed approach, read: [ http://en.wikipedia.org/wiki/New_API ], and for more information: [ http://www.linux-foundation.org/en/Net:NAPI ].

    21. Re:Make it Short and Fast and Snappy by gnuman99 · · Score: 1

      So, you want to create an SMP type cluster where each CPU is connected with another access a 100Gbps switched network? And your application sounds like it needs SMP to work but you want to go "on the cheap" and instead of getting a SMP supercomputer or even renting space in you, you just want to have a rack of playstations instead with 100Gbps network adapters?

      I suggest you fix your algorithm or get/rent a machine that can handle your data instead of guessing playstation will ship with 100Gbps ports for you anytime soon.

    22. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      Hmm, does IP over Infiniband run under Linux on a PC with PCI-e? Or any other <$5000 hardware?

      --

      --
      make install -not war

    23. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      PS: That post my 16,362nd. Probably fun to post #16384. My UID is (nearly exactly) 1/3 yours, so we were at equal rates when I posted #15313 - you're at 93.6% of mine. We both probably need a faster connection just to keep up with our posting rate ;).

      --

      --
      make install -not war

    24. Re:Make it Short and Fast and Snappy by macemoneta · · Score: 1

      The total bandwidth doesn't have to reach 100Gbps, but only exceed 10Gbps, to require faster than 10G-e, which only 8-10 SATA drives in parallel could do today.
      The SATA interface speed isn't of much use in this calculation. While SATA drives can burst to their interface speed when data is in the drive cache, they can only sustain 50-70MB/s (400-560Mb/s). In order to bring a 10G port to 80% sustained utilization, you would need 15 to 20 drives running at capacity (meaning, no seeks). In reality, with normal seeking and fragmentation, you would probably need twice that (about 40 SATA drives).
      --

      Can You Say Linux? I Knew That You Could.

    25. Re:Make it Short and Fast and Snappy by Anonymous Coward · · Score: 0

      a server single box can't really saturate multi Gbps - if you need it for some reason put multiple network cards in.

    26. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      40 SATA drives is no big deal. I've got over a dozen SATA drives running in my home, mostly to store my media collection. And I haven't even begun collecting video, because 750GB-1TB drives aren't that cheap yet. Once drives drop below $0.10:GB, and there are more HD video titles (especially if for download or TiVo), 40 SATA drives will be fairly common in many homes, to say nothing of the media servers that are servicing them.

      --

      --
      make install -not war

    27. Re:Make it Short and Fast and Snappy by afidel · · Score: 1
      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    28. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      The part you missed was where I said "the point is that there are already machines that can use that bandwidth". My TV's HDMI port already does 10Gbps. When I want to mix a couple PiP streams with incoming videophone and a peek at the front door for the pizza delivery, all in HD swapped around my house to different TVs to get someone to answer the door so I can stay on the phone (or some other common combination), that kind of bandwidth won't seem so exotic.

      Of course I want to go "on the cheap". Everyone wants to go "on the cheap": the lowest price possible (even if no lower than possible). Even before I get that kind of home media distribution, I do indeed have apps (like VoIP) that I'd like to invest $5000 on some Cell blade processor rather than $50K on just the networking.

      Bandwidth demands are exploding. Even telcos don't have 100Gbps over 2550Km yet, as the demo we're discussing demonstrates. "640K ought to be enough for anybody" should be warning enough. When "on the cheap" is near the limits of what what's available at any price, we need to hurry up and extend the horizon further. Or we'll soon fall off the edge.

      --

      --
      make install -not war

    29. Re:Make it Short and Fast and Snappy by Anonymous Coward · · Score: 0
      I've got over a dozen SATA drives running in my home, mostly to store my media collection. And I haven't even begun collecting video

      Crikey! How long would your total playlist last, a year? What's the boob-count in your pr0n collection? I've got two drives running and most of my collection is video. You, sir, have outdone me.

    30. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      I've got a very large collection of live music recorded (legitimately) by people in the audience, like Deadheads, over the past 30-50 years. And a solid percentage of that by bytes is video.

      A 180GB drive holds only 45 DVDs. When you're talking about hundreds of shows on video, you're talking about dozens of drives.

      It's not yet at the point where I need more than 1Gbps, but that's mainly because I don't have the SW quite running seamlessly for real home multimedia. But I'm still accumulating content, including a Blu-Ray drive (40GB movies, not 4GB). By the time >10Gbps ethernet is <$100 a port, I'll be able to justify it with my personal mediasphere.

      Yes, we do live in an age of miracles. "Fast, cheap and good", even it's all in the middle range of those parameters, is available, and the peaks available mean even the middle is quite nice. But that's the reason to push the peaks, to raise the middle.

      --

      --
      make install -not war

    31. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      I already described in this thread how a single host can saturate 10Gbps. And multiple cards have sync problems. De/muxing 50Gbps across 5-8 cards is a huge problem, in sync and just CPU power that's better off in an ASIC/DSP/FPGA in the network controller.

      --

      --
      make install -not war

    32. Re:Make it Short and Fast and Snappy by Doc+Ruby · · Score: 1

      That MHGA28-1TC is very cool. At about $800 for dual 20Gbps, that's $200 for the equivalent of 10Gbps, which costs $500 standalone. But a much more usable package, without de/muxing to get 20Gbps, and probably better ability to "bond" (or equivalent) across both 20Gbps ports for simulated 40Gbps. If only there were more competitors, it might be closer to $100 per each 10Gbps, which is cheaper than even standalone 1Gbps now, and of course much more integrated (lower power, more manageable, less HW...).

      Since the product is on the market, there's probably not that much demand right now, or there would be more competitors.

      Thanks for clarifying this entire discussion. I rarely learn this much from just whining on Slashdot :).

      --

      --
      make install -not war

    33. Re:Make it Short and Fast and Snappy by shentino · · Score: 1

      DDR using 102.4 gbps?

      Wow, I didn't know you could dance that fast.

    34. Re:Make it Short and Fast and Snappy by afidel · · Score: 1

      Qlogic makes cards too, their Silverstorm 7000 is similar to the MHGA28-1TC in that it supports dual 4x links, but because they are basically a duopoly there isn't much price competition. Actually it's quite similar to FC where there's Qlogic or Emulex.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    35. Re:Make it Short and Fast and Snappy by jpmattia · · Score: 1

      The problem is now at 1Gbps and 10Gbps in Ethernet technology, and is because the processor overloads with the amount of hardware interrupts.

      That's not the point of the article. 100 Gb/s technology is not being invented to hook up just 2 or even 100 computers to each other. The 164 wavelengths each carry a different 100 Gb/s stream. This is the type of technology you use when are trying to connect a chunk of Boston to a chunk of Baltimore.

      In order to process streams that fast, the first thing you do is demultiplex the streams into wide buses. For example, current 40 Gb/s streams are first demuxed into 16 x 2.5 Gb/s streams, which are then sent to a specialized processor to route the packets. (BTW, there are no interrupts at that level. It's all synchronous, and the processor is dedicated.)

      I will tell you that it is a thing of beauty to watch those things in action.

      And congrats to the folks at ALU for the result.

      JP

  17. Re:yeah, this would fill my hard drives in .03 sec by pizzutz · · Score: 1

    Well the optical transmission would go nearly the speed of light, but I'm not sure how fast the transmitter and detector can keep up with splitting and recombining 164 channels of data.
    I'm sure this is being looked at as a means of an intercontinental backbone, rather than something coming into your house.

    --
    GE/CS/IT d- s: a- C++++$ UL+++ P-- L++++ E W+++$ N+ o? K- w---() !O M- V- PS+ PE(++) Y+ PGP+++(+) t+++ !5 X++> R- t
  18. hurr by snarfies · · Score: 1

    Jeepers, that's really cool.

    Meanwhile, I can't even get FIOS service in Philadelphia, one of our major cities, despite my keen desire to purchase it.

    1. Re:hurr by DarkSarin · · Score: 1

      really, because I'm out in doylestown, and I have it. 5/5. and I'm very happy.

      of course I forgot to pay my bill for a couple months, so they aren't happy with me...but it is still working, so whatever.

      --
      "We don't know what we are doing, but we are doing it very carefully,..." Wherry, R.J. Personnel Psychology (1995)
  19. BROADBRAIN!?!? by Anonymous Coward · · Score: 0

    That's all well and good, but when can I get my BROADBRAIN connection?

    BTW, my submission capcha is "monogamy" WTF?

  20. wait a minute... by dnno · · Score: 1

    the title says 100 Gbps
    the article says approx. 16 Tbps
    and the last sentance says how close we are to creating 100 Gbps ethernet and describes how the terabit link was created using multiplexed signals at 100 Gbps.
    so what the heck am i missing, because im confused as hell

    --
    feh, lots of things are pointless, this one too
    1. Re:wait a minute... by Detritus · · Score: 1
      --
      Mea navis aericumbens anguillis abundat
    2. Re:wait a minute... by Anonymous Coward · · Score: 0

      I think what they are getting at is that as someone has created a long distance link that can transfer at 100Gbps on a single channel, they will now be able to attempt creating 100Gbps fibre optics for LANs.

      Someone correct me if I'm wrong though!

    3. Re:wait a minute... by burning-toast · · Score: 1

      The cable (fiber optic) has multiple lasers sending through it at once (multiplexing). So a single laser could operate at 100Gbps, but 164 of them at once... well...

      They just use different "colors" of lasers for each 100Gbps signal.

  21. Re:yeah, this would fill my hard drives in .03 sec by Anonymous Coward · · Score: 0

    ...real time simulations. Translation: Halo
  22. Re:yeah, this would fill my hard drives in .03 sec by Anonymous Coward · · Score: 0

    As the splitting combining is purely passive, it introduces negligable additionaly latency, say in the order of femto seconds.

  23. !Ethernet by Lars+Clausen · · Score: 1

    Part of the Ethernet spec is to wait 9.6 microseconds after the medium appears to be idle before sending, then resend if it collides. Light moves about 3 kilometres in that time. Making an Ethernet of 2550 km pratically guarantees nothing but collisions. So while this is a hunkin' heap of net, it's not Ethernet.

    1. Re:!Ethernet by funaho · · Score: 1

      Well in order to have collisions you need at least two transmitters on the same medium, and this is a point-to-point full duplex fiber connection; there can't be a collision because the two ends transmit on opposite fibers. It undoubtedly required some serious TCP/IP tweaking through to take advantage of that throughput.

    2. Re:!Ethernet by jpmattia · · Score: 1

      Running at 100G is hard enough, so these 100 GigE links run synchronously. And they are point-to-point, so no collisions.

      In ancient history: Ethernet was an asynchronous protocol that allowed collisions. Phone lines were run with a synchronous protocol (called SONET.) As ethernet got faster, the links became point-to-point and borrowed from SONET, which included running synchronously.

      (Also, not all Ethernet is 802.3 Your wireless is a good example: 802.11 a/b/g. 100GigE is in the IEEE HSSG; last I checked it had no 802 designation.)

    3. Re:!Ethernet by leomaro · · Score: 1

      >Part of the Ethernet spec is to wait 9.6 microseconds after the medium appears to be idle before sending, then resend if it collides. Light moves about 3 kilometres in that time. Making an Ethernet of 2550 km pratically guarantees nothing but collisions. So while this is a hunkin' heap of net, it's not Ethernet.

      In the specs of Gigabit Ethernet, there is a lot that has been taken out of Ethernet to avoid overload and get more data transfered, one of the things is the Collision Detection (the one that states how long you have to be idle, the backoff algorithm if there is a collision) in the protocol CSMA/CD (Carrier Sense Multiple Access with Collision Detection), At Gigabit speeds there is no way to think in a common channel (like old days Ethernet where existed coaxial cable), and so CD has no reason to be, also, the channels are now much (much) more reliable than before, so the specs take advantage of that too (and there are minimum quality requirements for the net to work properly).
      At Gigabit speeds you do not have traditional old days Ethernet, you have dedicated channels (switches act just like that, and now a switches are almost as cheap as hubs) or multiplexed ones and your Eth card does not need even to know its frames are being multiplexed with others. Data loss at those rates comes from other sources, not from collisions, but from things like EM interference (if you have a copper channel), or package discarding because the client can not handle that much data rate, a broken channel or an overflow of data going into a switch or router.

  24. Less than you might think by Tsar · · Score: 1

    What's that in Library-of-Congresses per fortnight? Assuming you want a copy of the LoC which will allow you to reconstruct the entire collection, you'll need scanned images of all the documents, decent MP3 (or ogg) files of the audio recordings and at least SD-quality copies of all the video. That would require something north of ten petabytes. At 1.64Tbps, that would work out to less than 25 LoC/fortnight--and that's assuming Mom doesn't pick up the kitchen phone and kill your connection.
  25. Fortune by Sigl · · Score: 1

    Hilarious... The fortune that I had at the bottom of the comments for this artcle:

    There is more to life than increasing its speed. -- Mahatma Gandhi

    Did someone plan that?

    1. Re:Fortune by burning-toast · · Score: 1

      - Did someone plan that?

      Here is mine:

      To doubt everything or to believe everything are two equally convenient solutions; both dispense with the necessity of reflection. -- H. Poincar'e

  26. In other news by BlueParrot · · Score: 1

    The movie industry just shat a brick.

  27. Reading stories like this are nothing more than by MrSnivvel · · Score: 1

    swift kicks in the groin, when all you get at the house is a "blazing" fast speed of 26.4 Kbps on dial up. Thank you Verizon, fucking thank you... ISDN is not even an option where I live.

  28. fucking retarted by Anonymous Coward · · Score: 0

    who gives a shit about these speed records, I could give a flying fuck less in these higher education bandwith pissing contests.

    bottom line is, here I am still sitting at comcast, throttling my shitty bandwith upload and download and can't get decent speeds at some times. will comcast every adopt the technology in the article? HELL NO

    fuck these speed pissing contents.

  29. So... by electricbern · · Score: 1

    Let me see if I got this right... They made 164 lights of different color blink very fast and very far. Why can't Slashdot summarize things better?

    --
    alias possession='chmod 666 satan && ls /dev > il && tail daemon.log'