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

 


Forgot your password?
Close
typodupeerror
×
Technology

Holy Grail "Opt-Chip" - 100GB/sec? 135

Posted by Hemos from the i-believe-it-when-I-see-it dept.
silicon_synapse writes, "ZDNET has a story about the new Opto-chip which can supposedly transfer data at 100GB per second. Yes, gigaBYTES. A two-hour digital movie could download in 1/20th of a second. The only problem is making the rest of the computer fast enough to take advantage of it. " The researchers are being published today in Science magazine and claim that the U.S. military will be using this as early as next summer. However, I think this is going to be another case of wait-and-see - the technology sounds a little too good - "spray on" application and such.
This discussion has been archived. No new comments can be posted.

Holy Grail "Opt-Chip" - 100GB/sec? More

Holy Grail "Opt-Chip" - 100GB/sec?

Comments Filter:
  • Slashdot has gotten itself into a worrying cycle of reporting the "latest technology" without spending one iota of time examining the moral and/or spiritual issues surrounding that technology. Dearly beloved, we are gathered here today to discuss technlogy that will allow the equivalent of a two-hour digital movie to be transferred in one-twentieth of a second (!) This should, at the very least, give us pause so that we may reflect for a moment on some of the tough issues that this technology brings up.

    The Internet is a double-edged sword. While it has brought a wonderful resource to many people, it has also exposed our children to material that they simply should not see. It is currently legal for homosexual groups, Islamic groups, feminist groups, etc. to put up material on the World Wide Web, where it can be viewed by anybody, including your children. This is disturbing enough. But red-blooded Americans ought to be frightened at the prospect of groups like this having unlimited bandwidth.

    Clearly, there are good uses for this bandwidth. For example, you child could download Charlton Heston's classic film The Ten Commandments in a fraction of a second (after paying for it, of course!) Imagine a world where you could go online with your sons and immediately be able to download demonstration videos for the latest assault rifles from Browning, Colt, or any one of our great gun makers. But that same technology could be used to serve up material about subjects that are morally unacceptable, such as lesbianism or liberalism. You would expect that companies such as Hewlett-Packard, which has a woman as its CEO, would quickly mobilize to use this technology to spread their hatred. Is this a good thing? No.

    What we need is a group of men to study this technology and figure out how to use it for the moral good. I envision a committee of legislators (such as Bob Barr, Trent Lott, Jesse Helms, and Tom DeLay) would get together and decide what American and foreign companies should have access to this technology and what companies do not need it. This way, we can still foster the technology and allow it to grow while at the same time ensuring that it is not used to turn our children to lesbianism.

  • Re:Hot Grits, coming up! (Score:1)

    by Anonymous Coward
    I was reading the RFC on Hot Grits yesterday, and from what I read hot grits can't be simulated at speeds over 100MB/S, unless they are cooled. Due to the fact that they are warm, they are ineffiecent and must be cooler to be transfered at 100GB/S
  • finally something that will keep up with the need.

    I wonder if this will make it into the race for the new internet backbones?
  • palstic - see http://news.bbc.co.uk/hi/english/sci/tech/newsid_7 04000/704324.stm

    for the laypersons guide.
  • palstic - see http://news.bbc.co.uk/hi/english/sci/tech/newsid_7 04000/704324.stm

    for the laypersons guide.
  • FWIW, I can move 1TB of zeros from /dev/zero to /dev/null in 39.771 seconds on my computer. That works out to 25.747 GB/s. So even if we did nothing but send zeros back and forth, the computer would need to be faster to reach the full potential of the network.

    For more useful applications, like watching pr0n, you would need an extremely quick bus (and video card). Although you wouldn't need to compress it (MPEG decompression is a bit costly), so you would gain there.
  • Sound has been for quite a while (probably for a few billion years, anyway), so finding a pre-sonal computer would be rather difficult. If they really are targetting this stuff at pre-sonal computers, I think they're making a big mistake, since they'll alienate a considerable amount of post-sonal computer owners.
  • Yeah, 100GB/sec over fibre, wonderful. And when will that fibre be wired into MY house? And who will do it?

    I guess it's great for the rest of the industry that has access to high speed lines, but here I am, sending this post over a 33.6 connection.

    No one will see fibre to their house for decades to come, so what difference does it make.
  • > you run the risk
    > of lack of compatibility and thus you can only play a nitch and thus
    > S&D kills you again

    Just because you don't pronounce niche (neesh) that way doesn't mean you should spell it wrong too. Other than that, you're unfortunately right. Sorry, but niche is one of my pet peeves.
    #define X(x,y) x##y
  • The world doesn't end at the 49th parallel, you know. I want some of that broadband to come my way, if you don't mind :) add Halifax, Ottawa, Vancouver (probably left out at least one important one. TO is close enought to Ottawa. I don't know what to do with Montreal :)

    Other than that, you've got one hell of an idea!
    #define X(x,y) x##y
  • I don't think anyone feels there's much of a "battle" anymore in this area. Organic materials (polymers, plastics and the like) show much better effects, work faster, are more damage resistant, cheaper, are easier to process, etc. The list of advantages actually is quite long.

    Inorganic crystals are currently used in devices but they're on the way out.

    -Geoff
  • > The article mentioned using this stuff to fab processors.

    I think they meant that you can make the modulators using current processor manufacturing technology. I'm not on the side of making actual devices, but I assume you make patterns by using a mask and depositing around it. So you're limited by your ability to make a good mask.

    IIRC, similar devices are already used in routers b/c even the crystalline devices currently used are faster than any other current technology.

    The military is interested in making optical-driven vehicles. Right now, you could make a weapon by disrupting the electrical signals between controls and the vehicle. (So suddenly you can't adjust your flaps on your nice F-16.) So they want to switch to optical pipes between the computer and the flaps so this won't happen.

    As far as flexibility or durability, it depends on the medium they're using. You'd need to see the Science article itself. I'd concentrate more on the durability side, which is very good. I don't know why you'd want to bend it.

    -Geoff
  • Could I spray this stuff on my eyes? Contacts? Glasses?

    Sounds like it would make studying much easier!

    I don't think my eyes are even doing 56K....
  • So you say living a life of sin and having faith in Jesus means you will go to heaven? So if I led the whole human race away from God but believed in god myself I would get to sit with God?

    But how can sacrificing anything (animal / son of god) take away my responsibility for my actions?
  • visible light and radio waves are all parts of the electromagnetic spectrum. Radio waves are lower frequencies and light is a higher frequency. A Photon is a quanta or package of electromagnetic energy. Photons travel as waves and arrive and depart as particles, hence wave / particle duality.

    As a photons travel at the same speed the frequency is a indication of the amount of energy in the package or quanta of of the electromagnetic energy. Higher frequency == more energy.
  • So organised Catholic religion is bad because it does not emphasise Christ, but organised Christian religion which has the same practices as Catholic religion is good because it does emphasise belief in Christ.

    I thought your earlier posts suggested that nothing can exist between a human and thier god and all that claimed to do so was bad.

    So as long a someone claims a devotion to Christ then they are not guilty of sin, no matter what they do as a person, because Jesus died upon a cross to absolve them of all responsability for thier actions?
  • If not I nominate you for the position of arch bigot. This would be an example of something I would not want my children to see without having a chance to expose it for the bullshit it is.
  • So what about organised Christian religion, are they guilty of the same sins?
  • Captain Pedant to the rescue

    Do you mean lower frequencies / longer wavelengths or higher frqencies / shorter wavelengths. Lower wavelengths and higher wavelengths don't mean much at all.

    Captain Pedant runs away from the lynch mob as no one likes a smart ass.
  • So you say that in order to find salvation for my soul I only need faith and faith alone. If i am the most despical monster on the planet it does not matter, only my faith does?
  • I wonder if the MPAA has just laid a dog egg in thier undergarments after hearing about this tech. :)
  • Think about this, if you plug that into your computer that would be the stupidest thing for anyone to ever do. now, make me a CSU/DSU/Switch with that... Now we're talking. imagine, a 24 port switch with that as the backbone across town or as a backbone to anything! Hell, my current 100Mbit fiber connections are fast but when my servers transfer 8 videos at once (3 inserters each fiber node, several video libraries) it can bog that down fast. Now, think globally.. we have 100mbit connections and a very few gigabit connections on the trans-atlantic fibers... Slap these in place and Voila! instant bottleneck remover! Now dont think of this as cheap bandwidth, the trans-atlantic cable operators will charge you X dollars for X bandwidth, it dont matter if they pay 30% of that in operating cost or 0.00001%, They're the only game in town. That's why we ran our own fibers and ignored the telcos...
  • Which may service a floor of computers.

  • Thanks for the updated link. It looks like an impressive device, however, the web page states:

    The NX64000 innovative switch fabric delivers 6.4 Tbps switch capacity per-chassis

    So while this box does route -- and probably pretty fast -- the 6.4 Tbps number they quote is in fact for switching between interfaces across the backplane. Putting layer3 intelligence on top of this will (most likely) slow things down.

    -B
  • The Lucent NX64000 can route up to 6.4 terabytes per second. IBM and Lucent, together, have achieved over 3 terabytes per fibre.

    Since the link you gave is broken, I can't verify this but I believe that the Lucent NX64000 can switch at this speed, not route. There is a huge difference. Try applying an access list to 6.4 terabits (not bytes) per second and process switch it. It doesn't sound too likely to me

    -B
  • A hard drive can't transfer at any where near that speed, so what's the use? Unless you run and store everything in the RAM, but I don't know, is that even possible? Is so much faster than physical storage to be worth it? About the only use I can think of for it is in dumb terminals, and who wants those?!
    It might be more useful to think that they might start putting drive controllers in parallel or drives in parallel or maybe even just the heads. Drives are cheap enough where this could be done and have a resonable cost for the speed (but not for the average user). This would increase the speed of the access on hard drives. Just because there isn't a use for it now doesn't mean that we can't think up a use for it in the future.
  • Forgot one question. Basically a CPU routes an electrical signal through a bunch of electronic pathways, yielding a digital result on another set of pins elsewhere.

    So even with the alleged 100GB transmission speeds, the signal has to travel a certain distance, and the distance it can travel is limited by the speed of light.

    When I was studying photonic chips about a year ago, the fastest CPU I could think of might have clocked out between 10-50 GHz.

    Am I missing anything here?

  • what are you talking about. current technology cant support this? do you think the technology on your desktop is as good as it gets.
  • could leavethe laser pumping continuously at one wavelength and modulate it with this stuff.

    Correct me if I'm wrong, but I got the impression that this new stuff is capable only of demodulating. It does the optics->electrons interface, but I saw nothing about it's ability to do the reverse electrons->optics.

    Kaa
  • From what I understand, this just takes care of the nasty fiber(optic) -> wire(electrons) interface problem. It does nothing to make the pipes fatter or to make computers run faster. So yes, it looks like a useful gadget, but the applications are limited (at least until I network my house with fiber).

    Kaa
  • I'm really impressed, you got an absolutely content free moderated up to a 5.

    You're paying ~$2000 for your average PC and you've already got supercomputer-level performance, although the supercomputer you're competing with is 10-15 years old. Hell my Palm III would put the top of the line 1950s mainframe to shame.

    If tomorrow Intel or AMD found that they could manufacture a 1Ghz bus at a reasonable price the next computer you buy would have it. But if you want today's supercomputer-level performance you buy... that's right a supercomputer. But you'll pay... right again supercomputer-level price for it.

    About the time this technology makes it way into the personal electronics market (if it ever does) you'll have a computer that can use all the bandwidth.

    drewish

    Now it's time to go meta-moderate and see if I can nail the people that moderated your message up.
  • How does all of this (new techonology, existing stuff from Lucent, et al) compare to terabit optical routing? (I have Sycamore Networks in mind, but would be interested in hearing about others as well.)

    The announced technology talks about 100Mbs, which is still too slow compared to an all-optical network, if I understand everything correctly.

  • Great article. Thanks for the link, I missed that back in March. Having the amount of bandwidth on tap that this article predicts would make some amazing changes not only to the telco industry, but to the entire computer industry and society as a whole.. wow

    mmm... bandwidth... &ltdrool&gt...

  • Yeah, Ice, read The Bible.

    It's all true. From the outset, The Bible may seem like any other book written by man. However, this book was penned by men, but written by God. Yes, that's right, this book has been certified as Truth by God. There have been thousands of religions over the course of humanity. Each of the peoples that follewed each of the non-Christian religions were certain that they understood the world and beyond. They were all wrong. Because they did not have a book certified by God. So forget yourself and all that you know and, for your sake, just follow the teachings of this book. You know it's all true. It's certified.
  • DeCSS (etc.) + Gnutella = The REAL Internet. Centralized control of media is for dinosaurs.

    We're just getting this thing started! Organic chips that you spray on stuff. Ethernet cards with a maximun bandwidth of 100/mb/sec, oh, um...wait a sec..

    (yes I know about gigabit ethernet, I've also seen the *actual* throughput)

    --
  • Why stop at LANs/WANs when you could have each gubbins in your computer with a fibre interface. Then you could do away with all those mucky circuit board tracks and just have fibre cables between your processor, memory and disk drives. Power leads to each component would also be needed of course.
  • What about wire interferance? At those speeds, you're sure to have a lot of collisions. You'd even have collisions INTERNALLY at those speeds! Forget network traffic...

    -------
    CAIMLAS

  • How do you know that ? Did God told it to you in person ? Unless He did, I don't see why you should be an authority on it. After all only He decides the rules, not some drunk Anonymous Coward on Slashdot.

  • Which is why multicast is important.

    At (what is hdtv?) 1900X1200x3x60 we could
    multicast 243 RAW video streams to each house.
    (410 Mb/sec/channel)

    Compressed, assuming 8:1 compression (a modest compression) would give us 87.5 Gb/s, and 243
    channels of HDTV per home. (not counting tha audio, but the audio bandwidth is negligible at these rates. (cuts it down to 86.6 Gb/s assuming cd-quality 5-channel sound/channel of hdtv)

    Not bad at all.

    So: Multicast is important!!
  • Which is another way of saying that there are fewer failure-modes for fibre-pipes vs. standard
    conductors.

    Another advantage of fibre that you didn't mention:

    Lower maintanance!
    For example: They have to check all the cables in
    an F-16 periodically to look at insulator wear.
    (because of short-circuits).

    Fibre problems are easier to diagnose, because a potential "short circuit" would be incredibly difficult to cause, and because when a connection is degraded, (i.e. fibre is cut/close to cut) it is easy to diagnose without having to open up panels...

  • This is hilarious. I just read the first 5 posts on this article. All ACs, all score 0, but all good posts. What is wrong with the moderators? Why are these posts at the same level as the grits guy? Read at -1 moderators! In my opinion it signals a big problem that the /. moderation system wasn't designed to handle. Now that /. was a much larger readerbase, not everyone will be a member, and not everyone will bother to make an account. Also with the hidiously increasing number of trolls, moderators are increasingly moving to reading at 1 or above. That leaves out a lot of ACs who have a good point, but either don't want to bother to create an account, or have an unpopular viewpoint and don't want to endanger their account. I suggest that the moderation system be changed to something so you can browse based on what it was moderated to, not just the score. So people you could browse by insightful, or AC, and ignore the troll or flamebait posts. This would also require moderator to more activly moderate down the trolls, but they seem more willing to do that than moderate up the ACs.
  • ...is that they've been talking about putting terabits through fibers for a while, and showing it off in lab trials and demo systems, while Ciena [ciena.com] is actually doing it in a commercial product and shipping it to customers who use it and depend on it to carry their data. Today.

    Don't get me wrong - Lucent has some very smart scientists and engineers, and some great technology comes out of the ol' Bell Labs, but Lucent also has a FUD machine at least as effective as Microsoft's - check out this article [lightreading.com] at LightReading [lightreading.com] for some of the dirt.

  • Im not entirely sure but if you look at the title of the paper at the science.com site:

    Low (Sub-1-Volt) Halfwave Voltage Polymeric Electro-optic Modulators Achieved by Controlling Chromophore Shape

    The paper itself is'nt available without a subscription but from that it looks more like modulation then the reverse.... that said you should never judge a book by its cover (or a paper by its title ;-)....

    StormChaser
  • From what I can see your right about the computers running faster but I think it will make the pipes bigger, the reason being that at the moment most optical systems use either mach-zender type modulator (which are pretty slow) or they actually change the current biasing the laser to modulate it.... The problem with that is that if you modulate the biasing of a laser it also tends to modulate its wavelength (theres a lot of research gone into changing a lasers output power while maintaing its wavelength stability but they are all far, far from perfect).... With this new system though you could leavethe laser pumping continuously at one wavelength and modulate it with this stuff.... That means the wavelength wont move around so much which means you can have more lasers pumping into the same fiber at closer spaced intervals which means you can have bigger bandwidths....

    The real beauty is that you can use the existing fiber that already been laid and the fiber ampifier that are already there and just invest in new modulator/demodulator pairs at either end and get a much bigger bandwidth....

    StormChaser
  • Personally I think broadcasts should be on their own dedicated line. Either via the air-waves or via dedicated bandwidth from our cable company.

    I'd really hate to have interrupts while watching a comedy just becase my son is downloading some patch to his game. But this is exactly what would happen with multi-casted HDTV over a shared network connection.

    I don't mind cable companies selling Internet access with the same fiber ( ideally this 100Gbps ) because they're going to dedicate bandwidth to their hundred some channels, and only sell net access for the remainder.

    Plus, it's a hell of a lot cheaper to have a company BUY cable explicitly then wire it to every room in the office building / school than to multicast up to 100G/s from the internet. Sure then you've got 2 fibers comming into your building instead of one, but the cable company fiber is almost free ( minus the monthy charge, and assuming they've got fiber in your neighborhood ) Just think about how much a T1 line costs now. And we're talking about buying general purpose internet access ( which just so happens that 100G/s is going to the same multi-casted locations regularly ). This is no cheap purchase today.. And the rewiring of the entire internet backbone ( plus the streets from the trunk to your building ) are not cheap by any means.

    I still do not like the idea of using general purpose hi-tech network connections for simple broadcasting. It's the idea of doing graphics processing in the general purpose CPU verses a dedicated, optimized seperate processor. Broadcasting belongs on their own seperate channel.

    Telephone lines, on the other-hand, I think should migrate their way onto the internet.. They're not dedicated lines, in fact, they're pretty damn close to TCP connections. They just need a little more reliability than the current internet phones. ( Internet II perhaps? ) The biggest advantage that I find from this is that we can consume the unused bandwidth from normal phone lines that we share on our OC-X connections. I'm sure telephone companies already do forms of digital compression and silent block removals before hitting the trunk, but still..

    As a disclaimer, I'm no expert in these fields.. These are simply opinions based on what I've read.
    -Michael
  • <i>maybe wiring CAT5 into every room of every house, same as electricty and phone?). </i>

    Why not? Does this sound so laughable to you?
  • CNN Headline News was talking about how there was a ongoing battle between the guys doing this stuff with plasitc and the guys doing this stuff with crystal.

    So is the Opt-Chip plastic or crystal?

  • how can you transfer your whole HD in under one second if you cannot read from it that fast?

  • Umm, shine a laser at a point on a film of this stuff and you get a magnetic domain? Potentially change characteristics by frequency? Gee you could even roll this stuff up and squish it into something resembling a cerebrum..

    Can anybody say FPGA real real fast?

    Sounds like this kind of material is going to get engineered into gates and wires much sooner than semiconductor tech will get there.

    Might give a new twist to the moniker "Flash Memory"..

    Military must be shitting they let the cat out of the bag.
  • Even now, the last mile (ie the connection between you and your ISP) is the bottleneck in any connection. My ISP has several OC-3s and T1s, but am I getting any benefit from it? No, I'm using a fscking 56k modem! Even if every major city had several 100 Gb/sec lines between each other, giving that bandwidth to the customer is going to be a bitch (maybe wiring CAT5 into every room of every house, same as electricty and phone?).
  • ISSCC: Transceiver developers creep toward 40 Gbits

    Source EETimes

    A link is here [eoenabled.com]

    So, we are moving to faster networks....2002 is a date tied to shipping product for oc768 by someone in one of the articles I read.
  • I don't think this will be implemented in any computer anytime soon, seeing what a waste that would be. I was thinking more along the lines of throwing it in a router or other high-speed net equipment that could ship stuff around nets like crazy. No point in throwing it next to your IDE hard drive and PCI bus :)

    Also, does the referred page load really crappy for anyone else using mozilla? ;)

    Mike Roberto (roberto@soul.apk.net [mailto]) - AOL IM: MicroBerto

  • http://cbc .ca/cgi-bin/templates/view.cgi?/news/2000/04/07/op tochip000407 [cbc.ca]

    Small article at the CBC website about said chip..
    .------------ - - -
    | big bad mr. frosty
    `------------ - - -
  • Scientific American magazine devoted a large article to just such a device ("Optical computer"). Don't know when exactly but it was in the early 1990's. Are there any Computer Engineer's in the house? BTW: this is my 1st post.
  • Damn what I could do with a 100 GB/s download speed. What are some of the things you'd do?
  • As with many technologies, especially that of communications related ones (which is most of them), the point of making advancements is so that they become so common place and so small and so fast that we no longer see it. It's just there, and is all around us. If you have such high bandwidth, you wouldn't be talking about "getting on the net". It would just be everywhere. We'd know it's there, but it'd be invisible because the networking hardware would be so small and the infrastructure so vast. We would not be aware of its speed, because everything would be instantaneous.
  • The speed quoted would require 40 Infiniband switched fabrics, plus 63 Ultra 160 SCSI striped disks with very fast cache memory, connected to each switched fabric to cope with this speed. For what purpose?
  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • But hey, 100GB/s over long distance fibre would be a really good thing

    How about 3.28 trillion bits per second down 300km of fibre? Wouldn't that be amazing?

    ...or maybe just old news:

    http://www.vnunet.com/News/601094

  • Argh! Typo again! I've got to cut down on the jellybeans. Try: this link [lucent.com].
  • The Lucent NX64000 [lucent.com] can route up to 6.4 terabytes per second. IBM and Lucent, together, have achieved over 3 terabytes per fibre.

    The number of routers you'd need to network up every household at this speed would be phenominal, as each chassis has only a very limited number of connections. On the other hand, if you built a national backbone from these, say at 128 Tbits (not unreasonable - 32-64 fibres would not be expensive, and enough routers to load-balance and direct the data would not be inordinate), then built metropoliton networks for each town and city, at the 2-3 Tbit level (again, well within the capacity of these devices and well within what could be sensibly installed, run and maintained), and finally ran 1 Gigabit lines to each house, you could have a genuine broadband network.

    (As a totally pointless exercise, I sketched out, for myself, a crude network map, based on this design, and costed it. The numbers look like a Windows serial number, but that's probably still pocket-money for Bill Gates.)

  • The fetch(tea()) operation is now delayed until after reading the article.

    There IS a way to get the full potential of this device, using existing PC technology.

    1. Buy a card which can work at 100 Gbytes/s
    2. Buy a large number of PCs
    3. Wait for it... ...Build a Beowulf Cluster!
    4. Dip the cluster in Liquid Nitrogen
    5. Overclock the PCs as far as you can
    6. Build a multiplexer/demultiplexer circuit, so that all the PCs can use the same card.
    7. Hope!

    It'd be great for 4-way C&C, or 20-way Netrek. (Who plays Quake anyway? :)

  • As I understand them, the docs say route. Try this link, instead. (Why they couldn't use Apache, with the spelling check on, I don't know. :)
  • From a quick look at the technical specs, it is routing IP packets (unlike so-called wavelength routers, which just switch DWDM wavelengths). They talk about doing WFQ (Weighted Fair Queuing) etc.

    All in all, this is a pretty impressive box - other boxes from Juniper and Cisco are already available but don't scale so high. For a good article on the benchmarketing involved in terabit routers, see http://www.lightreading.com/, which also has a great piece on how to succeed in an optical networking startup ('at all costs, stop your engineers from developing a product'...)

    Access lists are not very relevant on core routers - all use of access lists, e.g. for multiple-field classification (IP addresses, port numbers, IP Protocol, URL, etc.) should be done on the edge of the network, where the available CPU power per Mbps is much higher. This is part of the standard DiffServ model (details at http://www.ietf.org/html.charters/diffserv-charter .html) which is being implemented in Linux, BSD (via AltQ), Windows 2000, and most routers.

    The idea of DiffServ is that the edge routers do the hard classification, then encode this in the packet in the TOS byte in a 6-bit number called the DiffServ codepoint. From that point, all routers just have to check the TOS byte to see which queue to use for a packet, which is pretty fast and much easier to do in hardware. Linux has pretty good DiffServ support - see the Advanced Routing HOWTO for details, or the article in Linux Journal this month.
  • This sounds really REALLY REALLY good. To the extent that I thought I was having a April Fools Day flashback. My musings are all related to how quickly (if the report is true) it would take to transition the substance into useful digital technology.
    • It mentions that the substance can be "painted on". How would it be applied to create the type of etched patterns required for a microprocessor, ram chip, etc.?
    • Digital applications require on/off transistors. Does the fact that the substance has a really fast transmission rate mean anything in an of itself, or is there another fundamental discovery (how to make a transistor out of it) required?
    • What about bleed over/cross talk?
    • Switching speed?
    As you can see, I'm limited in my knowledge of anything but the basic electronics, but that's where everything starts. So I invite y'all to fill in the blanks, so I can sit back and learn from the commentary.
  • Does anyone else remember this [slashdot.org] article from a year ago about the first "photonic circuits" being developed by a company called Nanovation [nanovation.com]? The link to the story in that article is dead, but I found another one [currents.net]. Ever since then, I've been pondering how to get around the hideous bottleneck that modern memory is, and the only solution I've arrived at is to make an optical RAM equivalent based on similar technology. The only problem is that, I haven't been able to figure out a way of connecting it to the CPU without basically including it as an unupgradeable L2 or L3 cache with normal slow memory beyond it.

    Now, however, we have a device capable of nicely interfacing with an optical cable for 100 GB/sec speeds. This may be the interface we need for making "slots" for upgradable optical RAM. Cool! I can't wait to get to move to fully optical computers. This is the technology of the future. I just hope that the people creating these technologies will be willing to license their patents out to other companies.

  • This is hilarious. I just read the first 5 posts on this article. All ACs, all score 0, but all good posts. What is wrong with the moderators? Why are these posts at the same level as the grits guy? Read at -1 moderators! In my opinion it signals a big problem that the /. moderation system wasn't designed to handle. Now that /. was a much larger readerbase, not everyone will be a member, and not everyone will bother to make an account. Also with the hidiously increasing number of trolls, moderators are increasingly moving to reading at 1 or above. That leaves out a lot of ACs who have a good point, but either don't want to bother to create an account, or have an unpopular viewpoint and don't want to endanger their account. I suggest that the moderation system be changed to something so you can browse based on what it was moderated to, not just the score. So people you could browse by insightful, or AC, and ignore the troll or flamebait posts. This would also require moderator to more activly moderate down the trolls, but they are more willing to do that than moderate up the ACs.
  • Some of the cool stuff some researchers are doing is integrating a laser onto a normal ASIC.... That means that the connection from the chip to the rest of the world is optical rather then the usual bondwires and (relatively) slow electrical ones.... That means that if you have a chip with a 1GHz internal clock the whole system can move data around at the same speed - stick a few of the lasers in parallel on the same chip and youve got a 1GB/s connection with the rest of your system...

    Now all we need is a way of producing RAM and peripherals that keep match with the speed....
    StormChaser
  • 100 GB/s periphs are not useless in todays market. Just because your rusted-out Pacer can't handle the Autobahn doesn't mean it can't cruise along at 55. Say the bus could techically handle 400 MB/s. That's still three times faster than the realized throughput from a pair of Yellowfin Gig Ether, and ten times as useful!! A 100 GB/s limit is like money in the bank!!
  • Recently there was a web design contest [sylloge.com] that had a very interesting constraint. All entries had to be less than 5K, with no server side help (Note: The contest is closed now). There were over 1,000 entries and it really generated a lot of buzz. Many egos are at stake here! The FAQ indicates that there were five DHTML recreations of Space Invaders, four versions of Simon, and six 3d Maze games! All less than 5K. People can do more with less.

    The reason I mention this is that file size, download time, and bandwidth are critically important to all kinds of people. The contest I mention reflects a kind interesting return to the basics. What can you do with as little as possible? People of all types, from programmers to artists to system admins, actually want to do more with less, but they don't for a number of reasons. Argh!

    Perhaps the internet pipes that companies are building are less necessary than we are being told. Certainly the need would decrease if we could remove layer after layer of bloat.

    While some applications need the power, many don't. More and more features don't necessarily mean you're getting a better application. Quite the opposite is often true.

    Theory: More bandwidth is being requested because too people are lazy or because executives are too stupid to facilitate good coding and project management. The average geek certainly doesn't want to be lazy; the average geek is detail oriented and they want to kick ass. Small, useful applications are beatiful things. In any event, I wonder how much a pipe most people really would need to have if we were all able to be more effecient from day to day.

    John S. Rhodes
    Usability and a whole lot more...
    http://webword.com
  • "Damn what I could do with a 100 GB/s download speed. What are some of the things you'd do?"

    Isn't that just internal transfers? I mean you will not be able to download at 100GB/s with a 56k modem.
  • Gods - just imagine if you could actually get EXERCISE moving through those multiplayer RPG worlds, or perhaps playing those virtual fighting martial arts games. You'd have entire generations of Conans & Bruce Lees, who might have never left their house!
  • maybe wiring CAT5 into every room of every house, same as electricty and phone?).

    Why not? Does this sound so laughable to you?

    No, it sounds like a damn good idea. But a lot of people just aren't getting it yet. I (along with 3 other people) are going to be leasing a house starting May 1. It was just rebuilt from scratch. Is there CAT5 installed? No. Is there conduit so we can easily put it in? No. It's very very frustrating. We're looking at HomePNA stuff but I'm not sure if any of it works with Linux. I may end up having to get a seperate DSL line from the other 3. It's just a big pain in the ass. Arrrghh!!!
  • Actually, silica (glass) and sapphire make much better fiberoptics then any polymers do. It is true that Organic materials are easier to process then glass, but they provide too much optical interference to be useful for longer then a few feet. A single glass optic cable can run for miles with little or no quality loss.

    On optical chips, they are using tiny (<1um) cylinders made out of sapphire to bend light in a sharp 90 degree angles. This is part of what makes devices like this possible. In the past light had to be bent in long optic cables with a large radius, or be reflected. These cylinders, refract the light at 90 degrees and will hopefully make more "optic circuits" possible in the near future.

    I'm Finishing up my Masters Degree in Ceramics Engineering

    -
  • Actually, this could be the technology which actually makes these "back up your data on the net!" companies practical.

    As people have pointed out in previous threads, actually backing up modern disk capacities is a nightmare; my current machine has over 50 Gb of storage, and there's no consumer backup media that can realistically cope with that sort of capacity in any reasonable timeframe.

    Remote mounting an encrypted partition on a remote server could work quite nicely. In fact, with the sort of bandwidth that's being quoted here, you could quite happily boot off it, although the access times could be a pain.

    I'm still sceptical about the claim that this is more bandwidth than we can use, though; internet users have proven themselves to have an almost limitless capacity for pornography, and streaming DVDs would eat a lot of network... :)

  • I'm a little spaced out due to tremendous studying, which is coincidentally on a topic similar to this. My question is, "Didn't they basically just make fast, efficient transducer?" If the answer to that is, "Yeah", then we really don't have that much to get excited about.
    There's still the ultimate problem of increasing the speed of electrical processing. We've got optical routers, so this new material would basically just be used once the data is ready to be processed. And then it's back to transistor sizing, copper traces, barrier tunneling, and all that wicked stuff that nobody is ready to solve.
  • Comment removed based on user account deletion
  • Wow that's pretty slick. Unlike all the other people who've posted so far, I'm not going to complain about not being able to use this bandwith, instead I'm going to reflect on the sheer amount of porn that one could download it. I mean with ethernet I, I mean someone, has the leg up on 90% of the us population, but with this EVERyBODY could get their porn. New age of democracy is what I'd call it. It gets my thumbs up.
    One other thing, I am AMAZED that the porn industry doesn't sponspor more internet technology, according to my statistics 90% of the bandwith is used for this purpose alone. If only we could get like porn only aisles through the internet.

    sweetness.

    -dennis the kid.
  • If this could be brought down to the consumer level, and you have fiber to your flop, and the switches can keep up, and the backbone is there..

    And you have a system that can keep up with the byte firehose...

    Then we would finally have the bandwidth required to push enough information down the pipe to a system to create a true, functional, VRML interface. Get some gloves and a vision rig, munge up some software to do the data to image translation (which is being worked on for the Internet2 project) and let the games begin!

    Who'd care about flatscreen monitors then? Would carple tunnel be a thing of the past? Would it make information overload a true psychological problem? You decide!

  • I don't know if "more with less" is always a good mantra

    Back in eary March, there was a /. article on optical switches with a response from interiot who linked to an interesting article:

    The Coming of the Fibersphere [upenn.edu]

    It made a pretty convincing argument that when a resource becomes cheap enough, success will go to those who can waste the resource.

    Damon

    Work as if you don't need the money,
    Love as if you've never been hurt, and
    Dance as if no one's watching.
  • What is the relationship between photons and radio waves?

    The article stated that the device would be useful as a modulator converting signals from fiber optic cables or satellites. Aren't these two completely different problems? Satellite signals would need modulation through an antenna, and optical fiber signals would need modulation through an optical input.

    Does this logically mean that the chip has 3 input/outputs? And can switch effortlessly between any 3? What physical form would a modulator like this take?

    Also, the material can be sprayed on, but what method do they then use to etch circuits? Are there any scientists out there who can answer these questions?

    PS, please check out my new Privacy and Security forum at http://www.idmaweb.com [idmaweb.com].

  • Busses need to be wider, not just faster. Considering that the adaptation of 32 bit computing was hampered for years by a prevalent operating system manufacturer, it seems GNU/Linux may finally come into it's own as hardware architectures diverge from x86 standards. Given that 64bit processor support has been shoehorned into Linux for quite awhile now, Linux stands in an excellent position to foster competing cpu architectures, as opposed to differing brand names essentially presenting the same product. Bring on the cheap 64 and 128 bit processors, Linux can be compiled to adapt. Whats more, single chip multi-processors may be a little down the road. We seem to be at the threshold of a GNU era for hardware nerds. -Gary
  • Not really... it's just like fiber.. the thing can support pretty fast traffic, but noone has the capacity to feed it fast enough. Suddenly we have a fiber interface (this..thing...) that can go at fiber speeds... Now nobody can feed the interface fast enough :)

  • Need to see the Science article (Score:3)

    by ghutchis ( 7810 ) on Friday April 07, 2000 @05:39AM (#1145624) Homepage
    I'll preface my remarks by saying that I haven't read the Science article and the ZDnet article seems pretty short on scientific details. That said, this is one of my research topics, so I know a little bit about the area.

    First off, I've seen some questions about the quote of "spraying" onto a chip. There are a variety of techniques, but I'm guessing they're talking about spin-coating or CVD (Chemical Vapor Deposition), which are both used routinely in manufacturing.

    Secondly, these electro-optic devices use "second order nonlinear optics" (for all you physics geeks). Basically, people have been using crystalline modulators like lithium niobate for years, but they're very expensive and hard to make. So most of the research in the area has gone into making organic/polymer/self-assembled modulators. The idea is that you encase your chromophore molecule (the "active ingredient") in a polymer or other strong-film environment. Then you use this film in a waveguide and use it like a switch. The best mental picture would be a railroad switch--the electrical signal switches optical tracks for the optical beam.

    Without reading the published results, it's hard to know if this is really a breakthrough. My questions would be whether it's actually a new chromophore that's giving better results, a better preparation method, or something else. It sounds like they're making some change to the preparation of polymer devices, which are behind the self-assembled films many labs are making now.

    Suffice to say, the *real* revolution will come if anyone can get a usable third order NLO device. This would allow optical-optical switching.

    -Geoff
  • Stupid question time - what is the maximum switching frequency of a plain old phototransistor?

    Exotic technologies are very neat, but I'm wondering if more conventional technologies might already work.

    If phototransistor switching speed is comparable to ordinary transistor switching speed, you could probably build an optical transciever more cheaply by using closely packed frequency channels with bandwidth comparable to the switching speed, and a prism or diffraction grating to split them for parallel reading.

  • Re:Never fast enough... (Score:3)

    by GFD ( 57203 ) on Friday April 07, 2000 @09:37AM (#1145626)
    I think you are missing the point here. The Bell Labs benchmark was done at 40gbs a channel. This story is talking about 800gbs on one stream of light or channel. (Not that I think /. is very current on its news either...)

  • It's not about the math... (Score:3)

    by agravaine ( 66629 ) on Friday April 07, 2000 @08:15AM (#1145627)
    "The experimental GigaChannel Ethernet multiplexer combines up to eight independent gigabit Ethernet signals into a single 10 Gb/s signal
    stream."

    It sounds dopey, but it actually makes sense - what they are saying is that they can fit 8 Gigabit ethernet channels into a single OC-192 carrier (OC-192 is an industry standard, ~10Gbps SONET optical data rate) They can't fit 10, because there is overhead in any SONET stream, and they'd need extra overhead to split out the 1Gbps channels from one another. It seems like they ought to be able to fit 9 channels in there, if they really wanted to.

    A thought just occurred to me - they may not be TDM muxing the 8 signals at all, but rather they are saying that they can cram 8 1Gbps carrier signals into the same frequency range that would normally carry a single OC-192 carrier. This would make it easier (read: cheaper) to split out one channel to drop it out at it's destination without having to have expensive 10Gbps/1Gbps mux hardware at each terminus, and it is consistent with them needing to have guard bands [dead frequencies between the carriers so one signal doesn't stomp on the one next door] between the 8 carriers. The more I think about it, the more I'm convinced this must be what they are doing - the other way would be *way* too expensive.

    So, loopy as it sounds, fitting 8 Gigabit Ethernet channels into the 'space' of a 10Gbps optical channel makes perfect sense when taken in context.
  • I hate it when people talk about discovering more bandwidth that we can use ( or process ). When has this _ever_ been a problem in our history. My immediate reaction to this is that if the MPAA were to read that article, they'd be needlessly ultra paranoid about DeCSS. I say this because you and I will NOT be able to practically download entire DVD's any time soon. In order for us to have 100G / s bandwidth right now with that technology, it would have to be an isolated point to point network. Which means that you will probably be very familiar with the other end of the connection. The fear of DVD copying usually involves complete access to the entire web and a random end point for transmittion. The statistical likelihood of a desired DVD being on your other end are rather slim.

    Another use for this 100G network would be in an office situation but again, unless you've got a point to point star network, you're multiplexing someone's data which will reduce your bandwidth ( to well within a computer's processing capability ). Still, it's a hell of a lot more than what we have now.. But again, live video feeds are from static, known points ( mainly within your building, or between a finite number of known buildings ).

    So then, let's apply this technology to general publicly available internet connections. What do we have.. Raw bandwidth that will be soaking up by Linux distribution downloads, pr0n and general web-site traffic. Meaning if you put more bandwidth up, then the population will increase the volume of it's downloads to fill it. It's a basic trend that I'd be hard pressed to not call fact. And lets put this into perspective.. These 100G connections out on the internet are hardly going to be noticed over the existing high bandwidth lines where the routers are the slow point. Yes you can put high perf routers, and yes you'll eventually be able to maximally traffic this data, but your 56K modem or 1Mps cable network is not going to take advantage of it. And I'm doubting that we'll see home connections any time soon. Buisnesses that can afford such a connection are probably going to be saturating it. This is because it's probably not going to be cheap, and a business doesn't usually indulge in a technology for economic efficiency reasons.

    Will this help? Of course. Will it be great? Hell yeah. Will you realize any benifit? No. Because it's like a savings account interest rate when there's inflation. You may be getting 10% interest in your savings account, but if inflation is 15%, that doesn't get you anything more tomorrow than you have today ( you'll just be hurting less than if you only got 2% interest ).

    The biggest threat I see to the internet is video feeds ( hence my focus in this article ). If the public sees high bandwidth, they typically chant video, which, in my mind, if it ever comes to fruition then imagine the effect of thousands of homes leaving their internet TV connection on all night ( like we do our internet radio here at work ). This is just a rant, but it reflects my current paranoia about public bandwidth.

    -Michael
  • Now now now, don't forget yourself. If you realy wanted to increase bandwidth, on a board, you'd either need fiber chanels or massively wide busses. Both of which are rather expensive, not just for the motherboard, but for the periferal manufacturers. The whole driving force in the PC industry has been supply and demand. IBM pushed MCA back in the 386 days ( absolutely better than ISA ( and probably even better than EISA ) ), but obviously the world didn't stampeed to this technology ( just like they don't stampeed to Alpha's apparent superiority, even though they can technically still run the same programs ). Granted IBM was proprietary. But the point is that the PC industry _can't_ just supe up their systems, because that costs money, and S & D requires an equilibrium for profit maximization ( and in the PC world, that just about breaks you even ).

    There's also the case of compatibility. Even if you could produce as superior device in both performance _and_ price, you run the risk of lack of compatibility and thus you can only play a nitch and thus S&D kills you again.

    The reason the PC industry has been technologically advancing so quickly is because there has been competition for maximally compatible components that simply run faster. ( Plus a segmented market with some willing to pay premium, and a large majority demanding the cheapest ). If we didn't have that diverse market, we'd still be running a 486 class machine today. ( And Alpha's wouldn't be windows 3.5 compatible )

    -Michael
  • The switching speed of a transistor is measured as it's gain/bandwidth product: as the frequency goes up, the gain goes down. At gain==1, the transistor isn't doing you much good.


    That said, the GBP of a normal transistor is much higher than the GBP of a phototransistor, primarily because the base area (gate area, for MOS transistors) is much larger in a phototransistor than a regular transistor (My, what a big gate you have! The better to see you with, my dear!).


    The large gate area equals larger capacitance, which slows switching times down.

  • pins pins pins (and power ...) (Score:3)

    by taniwha ( 70410 ) on Friday April 07, 2000 @07:11AM (#1145631) Homepage Journal
    So you want a faster I/O bus? well then you have to do either (or both) of three things:
    • more pins (# bits/clock)
    • higher clock rates (greater bits/time)
    • move the bus on chip meaning you can't just plug in a new card
    The first two mean more power - power being (very) roughly proportional to the number of pins and how fast they are waggling - and as a result hotter chips. The more pins solution breaks down pretty fast - you can double the bandwidth by doubling the bus width only so many times before running into packaging problems - remember at high frequencies you need 1 power/ground pin for every 3-4 signal pins - also plug-in card with >64 data pins are probably impractical Bumping the clock rate while keeping the bus narrow seems to be the way some parts of the industry's going (1394, RamBus, the new fast USB etc).

    For most of the PC space the third option is probably going to be what you see - more integration - buses going away or being pushed on-chip meaning that the chances for plug-in high bandwidth goodies are virtually non-existant - instead you get what was chosen for you by the person who chose the chip when they put the motherboard together.

    Anyway the thing to remember TANSTAAFL - everything is a compromise.

  • Does this mean I can... (Score:3)

    by spiralx ( 97066 ) on Friday April 07, 2000 @04:44AM (#1145632)

    ... download a Jon Katz article without having to go for a cup of tea?

    But seriously, this sounds like a great technology, and one needed to implement the "Internet of the Future", whatever that may be, put it is only one technology out of a host which are required. Sure, in the short term this will give rise to improvements in data transmission, but until a series of other breakthroughs are made this won't reveal its true potential.

    So yeah, 100Gb/second is possible, but not for quite a while yet.

  • Ways of speeding up the RAM, at least. (Score:4)

    by Christopher Thomas ( 11717 ) on Friday April 07, 2000 @06:07AM (#1145633)
    Some of the cool stuff some researchers are doing is integrating a laser onto a normal ASIC....
    [...]
    Now all we need is a way of producing RAM and peripherals that keep match with the speed....

    For the RAM, at least, the answer is straightforward. Keep latency at its current range, but _heavily_ interleave RAM both on a bank level and a chip level. You now have RAM that can get 100 cache row requests and service all of them with a batch latency of 7 ns (or 5 ns or [etc]).

    This would let you, say, put 8 or 16 cores on a die without worrying about cache misses slowing you down (as long as you have a deep miss buffer).

    This would also be useful for transferring vast amounts of data with good locality in a known pattern (for instance, triangle or texture data) from RAM to a peripheral.

    This is probably what busses will look like in a decade or two, as it's much easier to eliminate cross-talk and interference on an optical bus than on an electrical one.

  • Lucent (Score:4)

    by roman_mir ( 125474 ) on Friday April 07, 2000 @04:44AM (#1145634) Homepage Journal
    Here is a research that is done at Lucent Technologies: [bell-labs.com]

    Instead of switching from optical wave to an electrical charge they use optical repeaters with mirrors and optical amplifiers.

    "The DWDM-ready GigaChannel has been demonstrated over 40 kilometers of standard single-mode fiber using WaveStar MetroPoint and also over Lucent's flagship long-reach product, the Wavestar OLS 400G, using multiple 80-kilometer fiber spans with online erbium-doped optical amplifiers and dispersion compensation."

    However it's only 10GB/s. Maybe they'll learn to do better than that.
    "The experimental GigaChannel Ethernet multiplexer combines up to eight independent gigabit Ethernet signals into a single 10 Gb/s signal stream, enabling switches, routers and servers to connect at 10 Gb/s in native Ethernet format without the need for protocol conversion. The prototype complies with today's IEEE Gigabit Ethernet standard."

  • Exemplifies the need for PC's to change (Score:5)

    by Stiletto ( 12066 ) on Friday April 07, 2000 @04:46AM (#1145635)

    IO performance has always been a problem with PC's. We've had PC's around for how long... and all we have to show for it is AGP 4X????

    While CPU horsepower has been following Moore's law pretty well, the PC world has lagged behind in terms of bus bandwidth. "100GB/sec" peripherals are useless when your bus runs at 133Mhz.

    Let's start pushing chipset and memory manufacturers to start putting out faster busses and memory subsystems, and then PC's will finally begin to approach supercomputer-level performance.
    ________________________________

  • terabit networking (Score:5)

    by crow ( 16139 ) on Friday April 07, 2000 @04:53AM (#1145636) Homepage Journal
    Terabit and faster networking isn't totally cutting edge anymore. Lucent is talking about sending many terabits per second over a single fiber.

    What is interesting is the ability to process packets at that speed. This chip is critical in converting that optical stream into an electronic stream. The other part is a CPU or multi-CPU architecture to process the data. I'm sure Cisco is very interested in this.

    So with Lucent figuring out how to send multiple terabits per second over a single fiber, this company able to convert those signals into electronic form, and hopefully soon Cisco being able to process and route data at those speeds, we'll soon be able to forget about bandwidth issues on the Internet. Or to be more precise, the bandwidth issues will become almost entirely limited to the link between consumers and their ISPs.

Slashdot Top Deals

Get hold of portable property. -- Charles Dickens, "Great Expectations"

Close