Photonic Switching to Boost Internet Speeds

Da Massive writes "Researchers at the University of Sydney have developed technology that could boost the throughput of existing networks 100-fold without costing the consumer any more, and it's all thanks to a scratch on a piece of glass. After four years of development, University of Sydney scientists say the Internet is set to become, on average, 60 times faster than existing networks. According to the Centre for Ultra-high bandwidth Devices for Optical Systems (CUDOS) at the University's School of Physics, the scratch will mean almost instantaneous, error-free and unlimited access to the Internet anywhere in the world."

Comments

Without costing the consumer any more?

[neokushan]

Ha! The technology might not cost much more, but ISP's will milk consumers for all they're worth.

Re:Without costing the consumer any more?

[Brian Gordon]

Not to mention that it's ridiculous to say the internet will be faster than existing networks.. the internet IS existing networks

Re:Without costing the consumer any more?

[Anonymous Coward]

Sadly (if you see it that way), this technology will probably never gain a large enough foothold for it to become truly widespread (because it's just so damn much to replace).

It seems like a pretty popular thing to do these days to announce some marvelous breakthrough in technology to get everyone excited -- and then never speak of it again.

Re:Without costing the consumer any more?

[thelamecamel]

Yep, that's almost it. Optical switches/routing would speed everything up by a lot. But an optical router is a long way off - one of the unsolved problems is buffering light to avoid packet collisions. At the moment, light can be slowed down quite a bit (actually the record is 17 m/s - slower than a cyclist), but everyone is still working out how to store it reliably for long enough to build a router. There are other problems as well.

The advance that has been made is about time division demultiplexing - sending 64 different 10Gbit signals down a single fibre, and then being able to separate out the individual signals at the end. They've been able to separate one of these out using a 5cm long waveguide, of a new material called chalcogenide (As2 S3) which is being researched heavily at CUDOS.

The Scratch

[Anonymous Coward]

I love it how in these news snippets there is never any explanation of the technology, but long descriptions about the wonderful changes it will do to the world.

Re:The Scratch

[BiggerIsBetter]

I love it how in these news snippets there is never any explanation of the technology, but long descriptions about the wonderful changes it will do to the world.

Back in my day we didn't call it a "news snippet", we called it a "press release".

Is it just me

[IceCreamGuy]

or does this article leave everyone else a little hungry in the "details" department? How does this mean "almost instantaneous, error-free and unlimited access to the Internet anywhere in the world...?" How will it not cost the consumer more? I feel like there's a story about breakthrough Tb switching tech every six months, and we haven't seen any of them deliver on these kinds of promises. They make it sound like you can just drop some glass in your existing switches and they magically become superpowered, whereas clearly if the technology ever actually matures to market we would be paying out the ass for these optics-enhanced switches and routers.

Re:Is it just me

[SpinyNorman]

In practical terms this means nothing for internet users since download speeds are invariably limited by the server or your local internet connection - not the backbone.

Even in a hypothetical situation where you have a fiber (e.g. FiOS) internet connection and were connected to a server that required you to have a photonic home router to keep up, you'd still be ultimately throttled by your NIC, mobo bus, or CPU.

Faster/cheaper switches may be exciting to backbone providers, but really that's about it.

Re:Speed

[Rob Kaper]

Not just that last mile is a bottleneck. For the majority of services (even and sometimes especially the popular ones) there are also severe bottlenecks on the hosting end, many of which have nothing to do with bandwidth and/or latency.

If any of the hops between (inclusive) you and the service has any capacity/speed problem, you'll notice it.

Re:In this house we respect the laws of physics

[Icarium]

Throughput != Latency

It has always amused me how commonly businesses play fast and loose with the meaning of the word 'speed' when it comes to internet connections. Yes, higher bandwidth will result in a 'faster' internet experience, but the data is not actually getting to you any faster - you're simply getting more of it at a time, so the webpage/download whatever completes in s shorter space of time.

You can drive faster than a truck, but if you're delivering more than your vehicle can carry, that slow ass truck is still going to complete the delivery in less time.

Argh, pet peeve, bad car analogy and all, brought about by years of listening to online gamers brag about how they've got the fastest connection and then crying when it makes no difference to thier gaming experience.

Anyway, the article is a bit light on details - can't quite make out if they're talking about increased bandwidth or increasing routing efficiency.

error-free?

[192939495969798999]

So is that error-free as in, a lot fewer dropped packets via pixie dust, or error-free as in it's so fast that you don't notice the dropped packets? I have a feeling if lightning hits the "magic glass" router, it will still screw up just like the current ones do when lightning hits them.

Re:pico/mega/terra

[Anonymous Coward]

It obviously hasn't occurred to anyone else so I'll say it.

Firstly you don't want a 50% duty cycle on a data switch. That would basically scatter your bits all over the world. You want to switch, then send some bits. So switching in a picosecond doesn't mean you'll switch *every* picosecond.

Secondly, maybe it takes the picosecond switch a microsecond to recover before it can switch again.

Or it may just be a bad piece of journalism.

BTW a picosecond is a million millionth of a second not a million billionth of a second. The latter is a femtosecond. It goes milli, micro, nano, pico, femto. pico = 10^-12. You can see how easy it is to be off by 3 orders of magnitude, then ... ;)

Re:Yeah?

[Anonymous Coward]

We need lots of articles with prominent claims of "boosting speeds and lowering prices" so that some of that thought may seep into the non-geek world so more and more people may start asking why their ISP charges so much. This could possibly catch the attention of politicians, who while often are bought by industry, need to maintain the appearance of fighting for the little guy.

Re:1 picosec/switch != 1 million switches/second

[shrikel]

Word probably autocorrected "million million" to just "million".