Terabit Ethernet Inches Closer To Reality

An anonymous reader writes "Researchers from Australia, Denmark, and China have combined efforts to show the feasibility of terabit-per-second Ethernet over fiber-optic cables. The solution involves a photonic chip that uses laser light for switching signals, and a form of the exotic material type, chalcogenide, or arsenic trisulfide."

Comments

no good

[gEvil (beta)]

I'm sorry. I'd like to be able to have my terabit ethernet runs over distances longer than a few inches.

Re:no good

[BadAnalogyGuy]

I only need inches, if you get my drift.

Re:no good

[Smidge207]

A "few" inches? Define "few." If few='3' then I agree with you, if few='3000000000000' then I think we've got something personal to hash out behind the wood-shed...

=Smidge=

Re:no good

[von_rick]

You can have a repeater every 3 inches. Simple.

Re:no good

[Anonymous Coward]

You can have a repeater every 3 inches. Simple

Re:no good

[MadnessASAP]

You can have a repeater every 3 inches. Simple

Re:no good

[Anonymous Coward]

You can have a repeater ever 3 inches. Simpl

Re:no good

[g8oz]

It's called research, genius. It's not a product announcement. This is Slashdot. If you want to drool over things you can't afford go read Engadget and the like.

Re:no good

[FishWithAHammer]

Obviously you forgot your "NO CARRIER".

Re:no good

[PTBarnum]

The article doesn't say how far they can send the terabit signal, only that the receiver requires 5 cm of fiber to split the signal into lower bandwidth pieces. Presumably the distance between sender and receiver is longer than that.

Re:no good

[El Jynx]

Hey, with the prices of repeaters these days, that doesn't actually sound too bad.. or.. waitaminit, these are those complicated fiber optic multiplex thingamajiggers, no? Ah, well. Plenty of IT companies going bust to buy them off of, too :P

Re:no good

[ebuck]

My setup seems to have a repeater every two inches. Should I get a CCSPE (Certified Cisco Slashdot Posting Expert) to fix this or is it safe to adjust the font settings myself?

Re:no good

[LordEd]

No, the last one was valid. 5-4-3 rule. You can have 5 network segments connected by 4 repeaters, where 3 of the network segments can have user connections.

Now, i could add the fact that you can have a repeater every 3 inches, but since there are already 3 active replies on the others, my message wou#@&^$*% NO CARRIER

Re:no good

[ebuck]

For such a high speed link, I think that a CSMA/CD technology is probably the wrong answer. Your "bubbles" in the network wire of collision screaming must be incredibly wasteful. But hey, we live in a field where waste is justified by the comparative cost of hardware upgrades over man hours.

I'd love to route Ethernet packets tunneled through an ATM link set up on this kind of bandwith, but somehow I don't think that solution requires cutting edge research.

Ethernet is good at doing what it does well, but the tuning require to make this truly effective might make the mess of jumbo packets look like child's play. I imagine that as the speeds increase, so will the issues. It's a remarkable achievement to push Ethernet this fast, but it seems that it's the equivalent of making your car travel 0.6c. At those speeds you have to wonder if you're using the right vehicle.

Re:no good

[elcorvax]

You can repeat the same joke every 3 posts . . .

Re:no good

[drinkypoo]

You can have a repeater every 3 inches. Simple

You might not be all wrong.

Right now the idea of using such a technology in a cable in a data center seems like madness (ho ho ho) but technology has a way of making the impossible commonplace, given enough time.

Re:no good

[Anonymous Coward]

no that was in the part just happened to be his catchpa....

Re:no good

[RiotingPacifist]

You can have a repeater ever 3 inches. Simply

Re:no good

[Anonymous Coward]

C-C-C-C-C-COMBO BREAKER!

Re:no good

[plague3106]

Doh... one of our repeaters is busted!

Re:no good

[petermgreen]

For such a high speed link, I think that a CSMA/CD technology is probably the wrong answer.
The modern way to build an ethernet network (at least the important parts of one) is to use switches and full-duplex point to point links. Full duplex links do not use CSMA/CD.

CSMA/CD is rarely used at gigabit (I don't think i've ever seen a gigabit hub) and isn't supported at all at 10 gigabit and above.

TFA and the /. summary poorly titled though, this is about a physical layer advancement. That advancement may eventually lead to terabit ethernet, it may also lead to other standards with similar perfomance.

Ethernet is good at doing what it does well
What ethernet has done well is maintain compatibility accross gnerations. If I have old equipment with a 10baseT controller I can plug it straight into a modern network. If I have even older equipment with a BNC or AUI connector I can still connect it to a modern network without too much fuss or expense.

Re:no good

[electrosoccertux]

I think you meant "NO REPEA

Re:no good

[Anonymous Coward]

@gEvil:
Ahh but they didn't say how many inches ;)

@BadAnalogyGuy
Ahh but they didn't say how many inches ;) ... errmm.. in your case, I guess its got a different context! :o

Re:no good

[Anonymous Coward]

I was once in charge of a network on a very limited budget--and with an insufficient number of fiber strands running between the buildings. One of the buildings led up to my main switch this way:

Cisco 3550G
    [1 gbps copper]
Cisco 2950
    [10 mbps copper]
10 mbps Raylan fiber hub from 1994
    [10 mbps fiber to a second building]
10 mbps Raylan fiber hub from 1994
    [10 mbps fiber to a third building]
10 mbps Raylan fiber hub from 1994
    [10 mbps fiber to a fourth building]
CentreCom 10 mbps hub with AUI fiber connector

Re:no good

[Hogwash McFly]

So would your wife.

Re:no good

[Anonymous Coward]

If you're having problems with drift you should switch to singlemode fiber.

Re:no good

[Anonymous Coward]

You cn hxve a?regeater eherb 3 indhbs. Sidpbe

Re:no good

[Anonymous Coward]

Checksum error

That's an aweful lot of porn.

[tjstork]

Not that I would ever use a terabit connection for porn... but uh, when's that coming out again?

Re:That's an aweful lot of porn.

[Wrexs0ul]

Not really. It's a well-known fact a lot of innovation is driven by the porn industry. This stuff is probably being sponsored by the Ultraporn coalition to put their digital media online.

Imagine streaming video so clear you can actually sense the actresses' emotional issues!

-Matt

Re:That's an aweful lot of porn.

[geekoid]

False.

'Porn' gets into bed with all technologies. When using hindsight it may seem there are a predictor or a driver, but remember hind sight is a lying bitch.

Pretty much all 'failed' technologies have porn. dead technologies really aren't remembered. Hindsight is the worse kind of bias.

Finally:
Yes, that pun was intentional.

Re:That's an aweful lot of porn.

[Anonymous Coward]

Clearly, you did not bother to click the accompanying link.

Re:That's an aweful lot of porn.

[StikyPad]

Imagine streaming video so clear you can actually sense the actresses' emotional issues!

I thought that's what prostitution was for?

Re:That's an aweful lot of porn.

[Anonymous Coward]

Imagine streaming video so clear you can actually sense the actresses' emotional issues!

That would be the end of porn. I don't even want to imagine the kind of fucked-up psyche that leads one to a career in porn.

Re:That's an aweful lot of porn.

[QuantumRiff]

Um, porn went VHS, instead of betamax. Porn went Blue-ray, instead of HD-DVD. Both of these happened when the competing standards were new, and both times, the side they picked one the format war, after they had decided to use that format.

Re:That's an aweful lot of porn.

[blueZ3]

Correlation is not causation.

Re:That's an aweful lot of porn.

[MoFoQ]

funny thing...I was thinking that exact thing when I first read it...that and...I need to get a bigger hard drive. Also..."damn it....I just upgraded my home network to gigabit"

either way...reminds me of that video on *tube...you know the "For PORN" song.

Re:That's an aweful lot of porn.

[Anonymous Coward]

Being someone who works at a porn company with multiple dedicated lines buried under the ocean, I can say this is very true. We test all the equipment we have to the limits.

I worked for a lot of mom and pop companies that thought they had problems.

We are pretty much a dedicated Foundry and Cisco debugging team.

When a single server gets over 10,000 hits a second (yes, second, not minute) - it tends to stress your equipment.

Times that by a few hundred servers and you get the idea.

I used to deal with simple PHP and Apache issues before. Performance? Was never an issue.

Now half our stuff is written in heavily optimized C, our kernels are heavily tweaked and even Squid isn't fast enough to keep up.

We even have our own custom caching software.

Re:That's an aweful lot of porn.

[Narpak]

Imagine streaming video so clear you can actually sense the actresses' emotional issues!

Not to mention all the scar-tissue from aesthetic surgery and skin ravaged by too much make-up. High-resolution could be a bit of a problem for some; thankfully they are making improvements to CGI every year.

Re:That's an aweful lot of porn.

[fbjon]

What's fucked up about a career in porn, other than the obvious?

Re:That's an aweful lot of porn.

[MBGMorden]

On a large enough sample set, correlation implies a relation other than chance, and thus should be investigated. Otherwise you can keep screaming "Correlation is not causation" at every piece of data every produced and try to claim that we can never claim results.

After all, if I state that "Each time a plant is deprived of water and sunlight it dies.", stating "Correlation is not causation" is complete nonsense. We've observed over a large enough sample set that yes, in this case correlation damn well IS causation. Effectively, your only argument here should be whether or not the sample size is large enough.

Re:That's an aweful lot of porn.

[geekoid]

Again, wrong. Learn you history before opening your mouth.

There was porn on betamax. It was a violation of an agreement with sony, but it did happen. Had Sony not had there dumb ass licensing agreements and fee's, there would have been more. Of course I could argue that is Sony did that, betamax would ahve won. Then you would be using your hind sight to say the port 'predicted' beta would win.

There is porn on HD-DVD.
I have both.

The porn industry gets it more then the main stream movie industry does as far as getting your entrainment out their.

Re:That's an aweful lot of porn.

[geekoid]

Lets see:
Heavy drug use is rampant AND expected.
People drawn to that industry are usually screwed up emotionally.

I'm not against the Porn industry, but it ahs problems.

That said, as it become more mainstream it will draw people less screwed up. What the industry really needs is a union.

Re:That's an aweful lot of porn.

[geekoid]

I decided that risk was to high. I am on break using my work computer.

Re:That's an aweful lot of porn.

[RiotingPacifist]

Prostitution, recommended by 9/10 doctors as less virus ridden than using IE for browsing porn.

Re:That's an aweful lot of porn.

[geekoid]

Correct.
In this case a simple look at the history of home media technology to see that porn is on everything, so it's not a predictor. Also, there wasn't enough porn being sold to prop up the VHS industry; which is what it would need to do to make a winner.
Yes a lot of porn is made, but so are a lot of other movies. Also, porn was not mainstream at all during the VHS/Betamax war. I mean, you still had to order it out of porn magazines, or go to that seedy store just outside of town.

I think the samples size of all media is large enough for this case.

I mean, is he saying if porn was never printed, we wouldn't ahve printing presses?

Re:That's an aweful lot of porn.

[Anonymous Coward]

That said, as it become more mainstream it will draw people less screwed up. What the industry really needs is a union.

Then we'd get a lot of movies called porn that wouldn't have any sex in them.

Re:That's an aweful lot of porn.

[drinkypoo]

The porn industry gets it more then the main stream movie industry does

I would imagine.

as far as getting your entrainment out their.

Out their what? Wait, maybe I don't want to know.

Re:That's an aweful lot of porn.

[skolima]

"ahs"? "ahve"? Is this some kind of a new meme that I'm missing?

Re:That's an aweful lot of porn.

[vamin]

The plant example is a little flawed because it is clear that one event follows the other. No one would ever say that the plant's death caused it to be deprived of sunlight and water.

"Correlation is not causation" isn't nonsense if you're talking about two events that don't have a clear hierarchy. Suppose that rather than porn causing blu-ray's success, porn went with blu-ray because it was more successful. I'm not knowledgeable about the porn/blu-ray relationship so one of these possibilities may be provably false. I just didn't want to have to come up with a car analogy, and the porn analogy was right there.

Re:That's an aweful lot of porn.

[bigstrat2003]

It's a Futurama reference (and the link redirects to the WP Futurama entry). The professor gets de-aged, and is 63. His reaction is, "63!? Aww... now I'll have to get a fake ID to rent ultra-porn..."

Re:That's an aweful lot of porn.

[Jamamala]

While the constant screaming of "correlation not causation" annoys me as well, it is still an important concept.

For example; I might claim that every time I put a plant in a cupboard, it dies. I could then claim that cupboards therefore kill plants.

This would not illustrate the whole picture. Whilst being in a cupboard kills the plant, it is not some intrinsic property of cupboards that causes the plants to die. Instead, there is a causal factor associated with being inside a cupboard; lack of sunlight and water. It is this that kills the plant.

(I know, I know, you could reduce the argument further into photosynthesis, but if I did that, I'd have very little idea of what I'm talking about.)

Re:That's an aweful lot of porn.

[Anonymous Coward]

I just used all 5 of my mod points to drop you on your various vapid posts.

Re:That's an aweful lot of porn.

[jonaskoelker]

Every single time the earth's mean temperature has been rising steadily over 400 years, pirates have died out. That's a 100%!

Re:That's an aweful lot of porn.

[mmontour]

That said, as it become more mainstream it will draw people less screwed up. What the industry really needs is a union.

Bad idea - you'd end up with all of the acting jobs going to the women with the most seniority.

Re:That's an aweful lot of porn.

[Sybert42]

Someone meta-moderated to remove the downmods? Hey, I moderated your sharks with lasers down too. Somebody cancel that too!

I pay for a subscription, and the fake departments are bad enough without this OVERRATED trash.

Re:That's an aweful lot of porn.

[Anonymous Coward]

Methinks you don't know the sort of people who enjoy gay midget porn, or other erotica niches, for that matter. I'd say that in the future, some weirdos would get off on stuff like the GP mentioned, but I'm sure they already exist.

Re:That's an aweful lot of porn.

[Anonymous Coward]

Correlation is not causation.

Just because 2 things are associated does not mean one CAUSES the other. They might be both caused by another factor.

Re:That's an aweful lot of porn.

[indi0144]

that will give some value to some stains on some keyboards..

Re:That's an aweful lot of porn.

[evilviper]

After all, if I state that "Each time a plant is deprived of water and sunlight it dies.", stating "Correlation is not causation" is complete nonsense.

That's not a correlation at all. That's a directly observed fact, based on a single (well, two, actually) variable in otherwise controlled conditions.

That's vastly different than saying, etc., plants start dying when birds migrate over-head, and trying to suggest there's a link.

If you're going to specifically deprive some variable from group A, and compare their results to group B, it's not just correlation, it's a scientific experiment.

Re:That's an aweful lot of porn.

[jsiren]

Care to give an example? From what I've read, pirates are alive and well.

Live piracy map.

Piracy map for 2008.

If you're sailing in the areas where the above maps have red markers, you may want to read this.

Excerpt:

Gulf of Aden:
Somali pirates are attacking vessels in the northern Somali coast in the Gulf of Aden. These pirates are firing automatic weapons and Rocket Propelled Grenades (RPG) in an attempt to board and hijack vessels. Once the attack is successful and the vessel hijacked, the pirates sail the vessel to the Somali coast and thereafter demand a ransom for the safe release of the vessel and crew.

Re:That's an aweful lot of porn.

[DarkOx]

Correlation is meaningful, it tells us we might want to look and see if causation is there. No matter how big your sample of dead plants gets until we do some more investigation and deepen our understanding of how plants live we could just as easily say:

Dead plants lead to drought and darkness.

Now yes any first grader could tell you thats backwards but you get the point. Does the popularity of a new medium lead to use by porn or does being used by porn make a new medium popular. I think you have to dig a little deeper to answer that.

Re:That's an aweful lot of porn.

[drsquare]

On a large enough sample set, correlation implies a relation other than chance

OK then, we can say that snow causes cold weather. After all, there's a correlation.

And by my count, VHS and Blu-ray are two formats. Is two a large enough sample set?

Re:That's an aweful lot of porn.

[jonbryce]

Yes, the plant might die because it is deprived of water and sunlight.

It is equally possible that that the water and sunlight stops because the plant is dead and no longer able to power them.

It is also possible that an external factor is causing both the sunlight and water to stop, and the plant to die.

You have to carry out additional research beyond establishing a correlation to find out which of the three it is.

Re:That's an aweful lot of porn.

[MBGMorden]

I didn't say that correlation implies causation. I said that on a large enough sample set, it implies a *relation other than chance*. Your statement is an example of that. Snow and doesn't cause cold weather, but it most certain IS related.

And no, two isn't a large enough sample set. I wasn't arguing that it was. I was merely pointing out the just throwing up "correlation is not causation" every time someone points out a correlation is nonsense, because correlation of events on meaningful sample sizes is a very, very useful thing for determining relationships (of which causation is included).

Lasers? Them's some fast sharks!

[Anonymous Coward]

This brings new meaning to the word "terabite"!

sweet

[_avs_007]

Now I can finally get started on building my holodeck.

Re:sweet

[RiotingPacifist]

Didn't you get the memo, our world is a hologram, making a holodeck inside it would be like googleing google.

What value?

[pig-power]

Tera ethernet... 5-25 gig monthly caps... "I used my monthly cap in 31.65 seconds..UH O..."

Re:What value?

[Spazztastic]

Tera ethernet... 5-25 gig monthly caps... "I used my monthly cap in 31.65 seconds..UH O..."

That would mean the telco companies actually decided to give us enough throughput. Sure, it'll work well on a LAN when they eventually deploy it, but unless if you have fiber coming to your house and all the way to where you're trying to grab that episode of Desperate Housewives from it will not go that fast. You also have to account for your neighbor who is addicted to porn and downloads it constantly seeding at 100% for days on end.

Re:What value?

[Shakrai]

You also have to account for your neighbor who is addicted to porn and downloads it constantly seeding at 100% for days on end.

Hey, don't talk about me like that when I'm not around ;)

Re:What value?

[lordharsha]

Not in Australia

Terabit ethernet, govt filter, dialup speeds

5-25 gig is more than enough

(That's probably a bit dramatic, but I think you get the point)

Re:What value?

[pushing-robot]

Actually, 1 Terabit/s = 125GByte/s. So that 25GB cap would take... Wait for it... 0.2 seconds of continuous downloading :)

Then again, by the time bandwidth like that is cheap enough for us, it'll be cheap for the telcos as well, and we'll probably be moaning about the petabyte bandwidth caps on our $20/mo plans.

Re:What value?

[Killer Orca]

Speaking of caps, any Comcast customers here who have run up against to or close to the ceiling?

I think if two companies are in the same market then one of them will eventually blink by doing actual network upgrades to get customers from the other company. Of course that's assuming they want to make money by doing work and not raising rates.

Re:What value?

[Spazztastic]

Speaking of caps, any Comcast customers here who have run up against to or close to the ceiling?

I think if two companies are in the same market then one of them will eventually blink by doing actual network upgrades to get customers from the other company. Of course that's assuming they want to make money by doing work and not raising rates.

I've never come close to hitting my cap and I am a "power user." I also don't have an HDTV or a computer to stream HD from.

Re:What value?

[Mister Whirly]

Well, it would only take .2 seconds if the sending server was serving you up packets that fast. Somehow I doubt you would get maximum throughput speed. I know I never hit the full 100 Mb speed of my network when connecting to a server on the net.

Re:What value?

[michrech]

You also have to account for your neighbor who is addicted to porn and downloads it constantly seeding at 100% for days on end.

Hey, don't talk about me like that when I'm not around ;)

Only two minutes from OP to reply -- You type pretty fast for only having one hand available!

Re:What value?

[geekoid]

But if you look, that whole sentence could be typed with just the left hand keys.

no, not really I just wanted t make you look~

Re:What value?

[RiotingPacifist]

ah but are we talking HDD GB or O/S GiB
1Tib could well fill 25GB of your hard drive in 0.18s, but an ISP would have to be a complete douche to use such units of measurement...

Re:What value?

[Anonymous Coward]

Sure, it'll work well on a LAN when they eventually deploy it, but unless if you have fiber coming to your house and all the way to where you're trying to grab that episode of Desperate Housewives from it will not go that fast. You also have to account for your neighbor who is addicted to porn and downloads it constantly seeding at 100% for days on end.

I live in northern Europe you insensitive clod!

Re:What value?

[bigstrat2003]

Simple solution: switch hands.

Re:What value?

[Atari400]

Finally I can instantaneously discover I am stuck at 99.9%.

Obligatory

[Anonymous Coward]

Think of the HD Porn! (we all know porn's the real reason for improving networking... ;-))

Re:Obligatory

[LUH 3418]

You obviously don't watch much, HD Porn is already available!

Star Trek solutions?

[Anonymous Coward]

"The solution involves a photonic chip that uses laser light for switching signals, and a form of the exotic material type, chalcogenide, or arsenic trisulfide."
Very Star-Trekkie
Yes, and by reversing the polarity of the anti-matter heisenberg compensators, we can flufenog the grapstitle.

Seriously, though... I enjoy any solution that uses "photonic" anything and "arsenic trisulfide" anything. Cool

Re:Star Trek solutions?

[von_rick]

...I enjoy any solution that uses "photonic" anything and "arsenic trisulfide" anything. Cool

Interestingly, it conjured an image in my mind that is a mix of baby-formula and pesticides.

Re:Star Trek solutions?

[ettlz]

Dude, Star Trek technology (from TNG onwards) was based on optronics, not photonics.

Re:Star Trek solutions?

[Ragzouken]

Optronics is a sub-field of photonics.

Re:Star Trek solutions?

[Mister Whirly]

Actually, that is a new Chinese product where the pesticide IS the baby formula. Now with more melamine flavor!

Re:Star Trek solutions?

[El Jynx]

In TrekSpeak, Pesticide sounds strangely Ferengi. But I also have to admit reading the two compounds and thinking: "Holy smyte, where can I get tickets to that!" Maybe the arsenic was a giveaway.

Re:Star Trek solutions?

[Mister Whirly]

I'm going out on a limb here, but I am guessing you have never gotten laid.

Re:Star Trek solutions?

[Anonymous Coward]

In Soviet Russia, grapstitle flufenogs YOU!

Re:Star Trek solutions?

[drinkypoo]

This is truly a gigantic step forwards for the Chinese people, who will no longer depend on Western companies and governments to poison them.

Re:Star Trek solutions?

[Mister Whirly]

Really? When was the last time that US toys imported into China caused Chinese kids to become sick? I must have missed that news.
What's next then? 15 minute breaks for the 8 year olds working in sweat shops over there? Or maybe free lead-based cookies in the employee break rooms?

Hint: if you are going to rip the US government, citing the Chinese government probably isn't going to help your case.

Re:Star Trek solutions?

[Anonymous Coward]

Interestingly, it conjured an image in my mind that is a mix of baby-formula and pesticides.

Great Scott! You know what you've just done man? You've found the answer to the looming overpopulation crisis! Give this man a Nobel Peace Prize!(tm)

Re:Star Trek solutions?

[drinkypoo]

Hint: if you are going to rip the US government, citing the Chinese government probably isn't going to help your case.

Who buys all that shit anyway? Sung to the tune of the Bad Religion song "You Are the Government".

Too early?

[NotPenny'sBoat]

Tbps ethernet seems a bit early. Correct me if I'm wrong but isn't the average read on a SATA somewhere around 5 Gbps?

Re:Too early?

[imamac]

I think 3 Gbps is closer to the norm.

Re:Too early?

[NotPenny'sBoat]

I mean, wouldn't that be like using a firehose as a straw? Tons of speed, but what could you actually do with it?

Re:Too early?

[norkakn]

10 Gbps is already normal in server rooms. OC-768 is in the wild at around 40 Gbs. 100 Gbs is definitely around in labs, but I'm not sure if any of it is retail yet.

SATA doesn't have to be very fast, because a single hard drive isn't very fast.

Re:Too early?

[fuzzyfuzzyfungus]

If you have, say, a bundle of fiber running across the Pacific that would cost you 9,334 bazillion dollars and a battle with the giant enemy crab just to upgrade; being able to increase its capacity just by upgrading the hardware on each end is a very attractive proposition. This applies, to a lesser degree, in all but short run situations.

This isn't exactly destined for workstations in the near future(heck, neither is 10GigE, and that is more or less commodity-off-the-shelf stuff by now); but there are applications where higher speed per fiber could well be desirable.

Re:Too early?

[LoRdTAW]

I suppose you forgot about internet back bone links. Terabit Ethernet should hopefully enable Tier 1 ISP's to provide really fat pipes to ISP's so we can finally get more bandwidth. The bigger the backbones the faster our broadband can be. Well at least that's my fantasy. 100mbit boradband should be cake walk with tubes that fat.

Re:Too early?

[norkakn]

Let 1000 people drink?

Re:Too early?

[NotPenny'sBoat]

This is true, I wasn't thinking of the back bones. Now I feel silly.

Re:Too early?

[Reapman]

This is no different then 10Gig Ethernet. Your not using this for Desktop, but in large ISP backbones to handle traffic. I'm sure this is years away from practical use even in there however.

Wake me up when Cisco offers 1TB Blades.

Re:Too early?

[Anonymous Coward]

You shouldn't feel too silly. The fiber used to run the cross-oceanic backbones has been around a while. It's not clear that any early introduction of Tbps over fiber will be able to run on those links.

Re:Too early?

[QuantumRiff]

The article talks about using DWDM to basically multiplex multiple 40Gbps wavelengths on the same fiber. Separating out the wavelengths at the other end is the part where the speed limitation seems to be. 40Gbps has been around for awhile, and so has DWDM.

Re:Too early?

[ouachiski]

It wouldn't be just replacing equipment at the ends, you have the repeaters to that would need to be upgraded.

Re:Too early?

[Glonoinha]

Sustained real world throughput for SATA drives is somewhere in the 500Mbps range - that's 60 Megabytes per second for single-threaded sustained reads or writes. Mix it up a little by having multiple applications access the drive at the same time and throughput can drop a full order of magnitude (in the range of 6 Megabytes per second.)

Given that, yes TerE is serious overkill for anything you are not already using (and continually saturating) GigE for. I'd say about the only situation where TerE would really help is for setting up a single dedicated machine with a MASSIVE shared ramdrive, and having other machines use that ramdrive as if it were local. Actually the more I consider it, that would be a damn fine use of TerE, and a good way to improve performance in computer clusters working on shared solution programming.

Re:Too early?

[jandrese]

Unless your shared storage solution was absurdly large TbE is still going to be overkill. Even 10GbE is difficult to sustain without half a rack full of machines. 40GbE is still considered overkill for pretty much anything outside of an internet backbone link. TbE is more bandwidth than you can handle.

Re:Too early?

[Anonymous Coward]

10GbE isn't all that common in server rooms at all. The price and compatibility issues make it prohibitive. 100GbE is still a long way off anything like an actual product. The best you'll do in a "LAN" at the moment is something like QDR Infiniband, which isn't cheap and is pretty specialised.

Yes I do work for a 10GbE vendor.

Re:Too early?

[Anonymous Coward]

Ok, so lets expand a bit here. Instead of a connection to 1 machine imagine this is the uplink sitting in front of a data-center with 5000 machines all sending individual streams to 500,000 people at once.

Re:Too early?

[Anonymous Coward]

No, with modern repeaters, that is not necessary. The repeaters are pure optical amplifiers that don't care how the signal is modulated. Only the end equipment needs to be sophisticated to do the fine-grained wave-division multiplexing - so you don't need to pull up the repeaters to upgrade the capacity. It's really quite neat. See http://en.wikipedia.org/wiki/Optical_amplifier

Re:Too early?

[Anonymous Coward]

Sustained real world throughput for SATA drives is somewhere in the 500Mbps range - that's 60 Megabytes per second for single-threaded sustained reads or writes.

Your data store needs an update.

Newer SATA drives can sustain 120 MB/sec streaming reads or writes.

Just don't seek.

Re:Too early?

[geekoid]

True, but you can slice it up to many customers using less cabling. SO it might be terabyte to your nod, but then everyone gets a paltry Gigabyte for the last 100 yards.

Re:Too early?

[Lord Ender]

I thought everybody had a striped RAID with 300 SATA disks these days. Is it just me?

At any rate, the idea (at least at first) would be that the switches and routers are all linked up with Tbps Ethernet. Then users hanging off of these with Gbps Ethernet could transmit to each other at full speed.

Re:Too early?

[Hyppy]

We have a full-heigh (~2 meter tall) enclosure full of Fibre Channel and SATA disks for a network attached storage appliance. I've never seen us use more than 5Gb/s in network throughput total.

Re:Too early?

[nikanj]

Fantasy? 100mbit broadband is quite common already, at least here in Finland :-)

Re:Too early?

[Glonoinha]

Agreed. Newer SATA drives can sustain burst peak throughput of 120MB/sec for the first 10% of the drive, on a newly formatted drive in which all the data is laid down in a single massive contiguous linear string of sequential blocks. During the first few minutes the drive is being used for benchmarking under optimal conditions.

Newer SATA drives that have been used in real life for a few weeks, with a normal distribution of data and the file allocation table scattered all over the drive, using the entire spread of drive from the fastest first few cylinders through the very last and slowest cylinders, average about 60MB/s sustained.

My ramdrive reads somewhere between 1GB/s and 1.4GB/s sustained, but unfortunately I can't make it big enough to actually do anything with it, and every time I power cycle the machine (which is often) I lose all the data on it. I'd consider buying a single dedicated server and putting 64G or so in it so a) it would be big enough do practical work with, and b) I could share it with all the machines - if only my network was fast enough to take advantage of it. 125MB/s is peak throughput for GigE, which is just barely faster than the average throughput of hard drives - in order to take advantage of that approach I would need something that can move 1.25GB/s sustained (hence my hypothetical interest in TerE, or at least 10GigE.)

Re:Too early?

[LoRdTAW]

Ah! You sir, are very lucky! I wish I lived there now, I love the SISU trucks!

The population of Finland is just over 5.2 million (so says the CIA world fact book). I live in New York City with a population of well over 18 million! Imagine if only 5% of the NYC population (~900,000) bought 100Mbit broadband. The peak traffic that could develop from that many people would be 90 terabits! And thats just one of our large cities in a large spread out country of just over 300 million. The fastest broadband I am currently aware of is Verizon's 50/20 Fios, a fiber to the curb tv/voice/internet system. You get 50 mbps down and 20 up. Problem is it is still very new and not yet deployed city wide as of yet (I am still waiting). Some cable companies are offering 10-30+ Mbps high speed packages and future DOCSIS standards will push that up around 100Mbps.

Re:Too early?

[Anonymous Coward]

Never underestimate the lack of spanning-tree in the wild, bridge-loops, and other acts of self-destruction.

Re:Too early?

[gatkinso]

Crab People... Crab People...

Re:Too early?

[Anonymous Coward]

100mbit is an old service in Sweden :-)

some places start to offer the 1Gbit ;-)

Re:Too early?

[norkakn]

10Gbps is cheap enough that even third tier colleges are thinking about using it for their trunks and inter-rack links.

Infiniband is solving a way different problem.

All-natural ingredients...

[geekmux]

"...a form of the exotic material type, chalcogenide, or arsenic trisulfide.

Whew, for a minute there I was worried we were going to use some hazardous materials.

Re:All-natural ingredients...

[von_rick]

But its got what networks crave, its got electrolytes.

Re:All-natural ingredients...

[Rayeth]

Actually materials like that are used in semiconductors all the time. Gallium-Arsenide is a popular doping material in fact. There is a lot more arsenic in your computer right now than you expect. As long as you're not eating it, everything is fine.

Re:All-natural ingredients...

[drinkypoo]

You laugh (me too) but I'd sure like to know what we're going to do with all the Arsenic we have lying around. I mention it often, but here in Lake County California we have a superfund site full of the stuff. If we could bind it up and then dope something with it that would be very stable, it might give us a future use for the stuff that would let us not dump it into a concrete pit, fill it up, then pave over the pit, build some new walls, and add more arsenic.*

* I don't know that they're actually doing this with Arsenic at The Geysers geothermal power plant, but they are definitely doing it with other toxics. The superfund site is out of Middletown, where Calpine's office is/was located.

Re:All-natural ingredients...

[Anonymous Coward]

But, the electrolytes don't actually *do* anything!!

Re:All-natural ingredients...

[Anonymous Coward]

Go away! 'Batin'!

Content Filtering

[El Torico]

Too bad my bullshit detector only operates at about 500 words per minute.

Re:Content Filtering

[Anonymous Coward]

On the flip side, your bullshit creator operates at the speed of light.

usefull for offline storage...

[gandhi_2]

...of the entire internets. just right click the network icon, select "save as" and name the file. Wait 30 seconds for the entire internets to download.

Re:usefull for offline storage...

[Klaus_1250]

Sorry to ruin the joke, but on a serious note: With 1 Terabit/sec and 30 seconds, you might be able to download a millionth of the Internet.

Re:usefull for offline storage...

[geekoid]

So your saying the entire internet can be download in 347.22 days?
Not too bad.

Re:usefull for offline storage...

[Klaus_1250]

Given the daily growth of the net, no.

Re:usefull for offline storage...

[Xtravar]

...of the entire internets. just right click the network icon

What, I thought the blue E icon was the internet! I don't get it!

Toxicity Potential

[Anonymous Coward]

Aren't Arsenic compounds toxic? And aren't we (globally) trying to move away from exposing the environment to yet more toxins?

I am assuming that the photonic chip uses crystalline Arsenic Trisulfide. (http://en.wikipedia.org/wiki/Arsenic_trisulfide) "Crystalline As2S3 however tends to oxidize on the surface, forming a layer of toxic arsenic trioxide."

I would love to hear from those more knowledgeable about these kinds of substances to understand if there would be any potential threat to the environment.

Re:Toxicity Potential

[nicodoggie]

Well, this tech certainly solves this guy's rat problem.

Re:Toxicity Potential

[Soft]

Aren't Arsenic compounds toxic? And aren't we (globally) trying to move away from exposing the environment to yet more toxins?

Compounds, no. Virtually all the lasers you are using are made of gallium arsenide.

Feasable?

[mail2345]

The article never mentioned the economic feasibility.
I highly doubt that for now that this will be cheap.
How much does this arsenic trisulfide stuff cost, anyway?

Re:Feasable?

[baldusi]

The article actually says that they've developed a process to make it on CMOS factories, although with different materials. So the potential is there to relatively "cheap" solution. Just remember that this is for backbone applications, where you count cost at thousands or tens of thousands per port.

Re:Feasable?

[markov_chain]

My question was how they are going to get permission to mount this stuff on fire hydrants? Seems pretty expensive.

My Understanding

[baldusi]

The interesting part is that they have developed a material to reduce the "fiber" needed to demultiplex the signals from several meters to just 5cm (~2in.). And a process to easily manufacture that. Apparently this would require extreme parallelism since each drop could only handle 10Gbps. I would be very interested if someone coul explain the particularities of Optical Time Division Multiplexing. I failed to foun any references and thus I'm no aware of the difference with simple TDM.

Still needs work

[smooth wombat]

The solution involves a photonic chip that uses laser light for switching signals, and a form of the exotic material type, chalcogenide, or arsenic trisulfide."

Once you have the photonic chip installed, you will need to realign the deflector shield to output a graviton pulse through the arsenic trisulfide to create an anti-tachyon pulse which will modulate itself based upon the resonant frequency of the transport medium, thus allowing for longer distance transmittal of data than is currently possible.

Granted, it will take 15 years and research team of a hundred to complete, but it is doable.

Re:Still needs work

[ionix5891]

Seven is that you?

Re:Still needs work

[Rick Bentley]

Once you have the photonic chip installed, you will need to realign the deflector shield to output a graviton pulse through the arsenic trisulfide to create an anti-tachyon pulse which will modulate itself based upon the resonant frequency of the transport medium, thus allowing for longer distance transmittal of data than is currently possible.

Granted, it will take 15 years and research team of a hundred to complete, but it is doable.

No, they can do it in the middle of a pitched battle and before the next commercial.

Re:Still needs work

[A. B3ttik]

But they'll forget about it within a week.

Re:Still needs work

[Ragzouken]

Keep dreaming, it's La Forge.

Re:Still needs work

[eggboard]

Okay, I'm the author of the Ars Technica piece, and that make me laugh.

Talking to the researcher, Eggleton, made my head slightly explode, because he's looking 5 to 20 years into the future with the research he's on top of today.

But they have practical devices that show that the stuff can be hand-built, and that's what blows my mind.

The future isn't in plastics -- it's in glass!

Re:Still needs work

[drinkypoo]

The future isn't in plastics -- it's in glass!

You almost had me for a minute, but then I got to thinking, what do you think they're going to ship all that glass in?

Re:Still needs work

[eggboard]

This is a funny day at Slashdot. +3 guffaw points.

Actually, I was thinking chalcogenide could be a good new name for a mixed drink. Maybe grenadine, liquid oxygen, and something fizzy.

Re:Still needs work

[Eil]

Granted, it will take 15 years and research team of a hundred to complete, but it is doable.

See that Borg ship out there? You have 5 minutes if you don't want to be a member of its crew.

Re:Still needs work

[Anonymous Coward]

You have ten minutes, Mr. La Forge. Picard out.

Re:Still needs work

[snspdaarf]

Fizzy, like Alka-Seltzer?

Re:Still needs work

[Savantissimo]

"Chalcogenide" sounds more like an alien war crime to me -

"You stand or sit or squish whatever it is you're doing with those tentacles,
ahem, you stand charged with 1st degree Chalcogenide of a cixz of innocent Glugyws. How do you plead?

"Gleeble poot zoooom pop! Zorn digqsstdfft pop!"

[Judge bangs photonic ultra-gavel]

"Don't try to diffract me! I don't care whether their Abbe numbers were lower than your specification! You are accused of having them Schott! No spreading dispersions on the victim's indexes will be tolerated, even in in-camera proceedings!"

It would also appear...

[Brain-Fu]

...that we are inching our way towards the metric system.

Do The Math,..

[GHynson]

And the cost to wire a typical house? TbE cards = $30k each F/O Cable = $325k ISP F/O = $12k per Month Total = $367k plus your first born. No thanks, I'll stick to my 10MbE ISP. for $25 month

Re:Do The Math,..

[Ragzouken]

It's a pity that technology like this never gets cheaper.

Re:Do The Math,..

[lewiscr]

10 years ago, you would've said the same thing about GigE. Now I can buy Cat5e cables and a US$20 switch, and do the whole house for under under a US$100.

Re:Do The Math,..

[Anonymous Coward]

10 years ago, you would've said the same thing about GigE. Now I can buy Cat5e cables and a US$20 switch, and do the whole house for under under a US$100.

Fibre has been around for 20+ years,
And why is it still priced at arm and leg prices?

The kind of networking I want.

[Anonymous Coward]

I will explain what kind of networking I want. But in order to do so, I will first have to define some math operations.
 
Astute readers know what addition is, and that multiplication is repeated addition and exponentiation is repeated multiplication. There is an additional operation, less well known, called tetration, which is repeated exponentiation.
 
I do not understand why, when mathematicians often want to define something in the most general form possible, nobody has, yet, developed a universal operator that allows you to take this repetition to any level you would like. This is why I am about to define such an operator.
 
Suppose that [1] is taken to mean addition, [2] is taken to mean multiplication, [3] is taken to mean exponentiation, [4] is taken to mean tetration. Now we can stick any integer greater or equal to 1 inside square brackets and use that as our operator. So:

• 5 [1] 2 = 5 + 2 = 7
• 5 [2] 2 = 5 * 2 = 10
• 5 [3] 2 = 5^2 = 25

. The value of 5 [29] 2 is obvious and left as an exercise for the reader.
 
Those familiar with mathematics know about factorials, denoted by n!. This is the product of all integers less than or equal to n. Now that we have introduced all the prerequisites, let's talk about what kind of networking I want.
 
I want googolplex! [googolplex!] googolplex! yottabytes per planck unit of time. When you can give me networking like that, I'll be happy.

That's one bit every trillionth of a second.

[JoeD]

In one trillionth of a second, light travels .3 millimeters.

So the receiver has to be able to not only detect that bit, but process it in time for the next bit that's right behind it.

Pretty impressive.

Re:That's one bit every trillionth of a second.

[Anonymous Coward]

No, a single optical fiber can carry many wavelengths of light simultaneously. The whole point of this research is to make it easier to combine them at one end and separate them at the other end.

dom

I just upgraded to gigabit ethernet last year.

[crovira]

My CPUs, displays, hard drives and network keep getting faster, getter, faster, stronger, (no apologies to "Datf Punk",) but this is one hell of a jump in performance.

Damn.

At least I can't upgrade my eyes, brains (and right hand. :-)

I'm a leftie

[MoFoQ]

but I'm a leftie.

That makes it better.

[tjstork]

Imagine streaming video so clear you can actually sense the actresses' emotional issues

That would actually make it better. Some broad ruins her life completely to make me happy for a few minutes. Kinda balances out marriage.

Re:That makes it better.

[Anonymous Coward]

The lapdance is always better when the stripper is crying.

Re:That makes it better.

[tjstork]

The lapdance is always better when the stripper is crying.

My wife actually introduced me to that song. She's pretty awesome.

Re:That makes it better.

[Mister Whirly]

Was she crying at the time, to make it better for you?

Isn't it interesting

[bratwiz]

Isn't it interesting that China, one of the net's biggest censorship proponents is out there on the forefront of high-speed technology.

RIAA is training rats to chew cables

[140Mandak262Jamuna]

In a desperate attempt to put an end to file sharing by users, legal or not, RIAA has started a program to train rats to chew the connection cables and let them loose all over Europe.

Re:RIAA is training rats to chew cables

[missvolare]

Send this to the onion! I guffawed--it's great!

hath the duplex

[Lotharjade]

I think I would just be happy for my work to move up to 100 Mbs full duplex. Sigh...

Multiprocessing?

[HiThere]

The place where I immediately saw this being applied was in multiprocessor systems. Short distance. Admittedly, 3 inches is still a bit short, but was that mentioned as a transmission distance limit? I don't think so.

This might make a dynamite system bus for a multi-computer system. It would probably reach between motherboards. It may not really be "Infinilink", but then neither was the bus that was given that name.

fat pipes

[snspdaarf]

Ted Stevens, your internet is ready.

What's the use

[dimethylxanthine]

when the Pirate Bay is about to go down and Australia has just started filtering Bit Torrent?

Imagine

[Anonymous Coward]

Imagine what this could mean for all the previously imagined Beowulf clusters...

if only...

[Anonymous Coward]

if only we could build a bridge to close the rest of the gap to this terabit-thea.

Another misleading summary...

[Wdi]

This material description is totally misleading.

From Wikipedia:

A chalcogenide is a chemical compound consisting of at least one chalcogen ion and at least one more electropositive element. Although all group 16 elements of the periodic table are defined as chalcogens, the term is more commonly reserved for sulfides, selenides, and tellurides, rather than oxides.

There are tons of different chalcogenides, and the arsenic component is not the defining element (here, it is sulphur).

I'm still waiting for my cheap 10gig-e

[Anonymous Coward]

Whatever, this is all very nice research, but I'm still waiting on inexpensive 10 gigabit ethernet.

The throughput of an inexpensive NAS is now bottle necked by 1 gigabit, it's about time it got cheap. How about some research on that?