World's Fastest Fibre Line Can Support An Internet's Worth Of Data

World's Fastest Fibre Line Can Support an Internet's Worth of Data

Currently, the fastest commercially available fibre optic line tops out at 100Gbps. That's super fast, sure, but isn't nearly a wide enough pipeline for our increasingly interconnected systems. That's why this new, multi-modal, fibre line is so exciting — it can pack 2550 times as much data into the same glass strand.

That's right, a 255 terabit-per-second throughput. That's enough to transfer a 1GB file in 31 microseconds (.003 sec) or move the entire contents of a 1TB drive in three tenths of a second. In fact, 255 Tbps is roughly the equivalent to the peak load on the Internet's transatlantic backbone. This single strand could, quite literally, replace the existing plethora of single-mode sub-sea cables.

The record-breaking technology was developed by an international team of researchers from Eindhoven University of Technology in the Netherlands and the and University of Central Florida here in the US. They were able to achieve this impressive feat thanks to an existing (but prohibitively expensive) technology known as multi-mode fibre. The existing transatlantic cables are all made of single-mode fibre — that is, each line can only carry data from a single source laser — but this new multi-mode fibre contains seven separate "cores" and can therefore each multi-core fibre carry up to seven distinct signals simultaneously.

The team also employed some tricky data transmission manipulation to pack the line. They first leveraged a spacial multiplexing (SM) technique — where individual encoded data signals from multiple sources are transmitted in parallel — to reach a speed of 5.1 terabits per carrier. They also utilized wavelength division multiplexing (WDM) — which works similarly to SM but separates and transmits individual data streams in varying wavelengths of light — to get 50 separate signals moving down the kilometer-long line simultaneously.

In short: it's a mix of hardware and software innovation.

As the team's research abstract in the October issue of Nature explains:

Single-mode fibres with low loss and a large transmission bandwidth are a key enabler for long-haul high-speed optical communication and form the backbone of our information-driven society. However, we are on the verge of reaching the fundamental limit of single-mode fibre transmission capacity. Therefore, a new means to increase the transmission capacity of optical fibre is essential to avoid a capacity crunch. Here, by employing few-mode multicore fibre, compact three-dimensional waveguide multiplexers and energy-efficient frequency-domain multiple-input multiple-output equalization, we demonstrate the viability of spatial multiplexing to reach a data rate of 5.1 Tbit s−1 carrier−1 (net 4 Tbit s−1 carrier−1) on a single wavelength over a single fibre. Furthermore, by combining this approach with wavelength division multiplexing with 50 wavelength carriers on a dense 50 GHz grid, a gross transmission throughput of 255 Tbit s−1 (net 200 Tbit s−1) over a 1km fibre link is achieved.

Obviously, this technology is far from being field-ready. Not only would replacing the existing single-mode infrastructure with these more-expensive multi-mode lines be expensive and difficult; it'd require entirely new routing hardware as well. It'd be like replacing every two-lane highway in America with eight-lane raised freeways. It's totally possible, just not economically feasible. But, given how fast the internet is currently growing, it won't be long until we need this technology. [Nature via Extremetech]

Picture: Sukharevskyy Dmytro (nevodka)


    This is fantastic, great use of technology.

    However, how easy is it to tie in to existing technologies?

    It's literally impossible to REPLACE the existing infrastructure, because doing so would cause an outage, and even a momentary outage to flick the switch on the backbone of the internet like that would cause disasters.

    The new routers that accept this multimode fiber need to be able to plug into existing routers. if they don't the technology will always just be a good idea. Take a look at IPv6 for a real world demonstration of what I'm talking about.

      ... You would replace it by laying the new cable then eventually retiring or re-purposing the old, perfectly usable infrastructure.

      And no, obviously it's entirely incompatible with existing infrastructure but could be decoupled INTO existing infrastructure similar to how FTTN works in many ways.

      There's no reason this will be 'just a good idea'... And will be adopted for core infrastructure the same way traditional multimode was..

    There is no doubt in anyone's mind that Optic Fiber and light are the future of computing. Internal computer components using traditional binary-Turing logic just using light would be the single biggest leap in computing since the creation of the transistor.

    ... over 1km, not the Atlantic. So not really the same.

      ... Their test was over 1km.. There is no fundamental difference over distance with fiber..

      There is always attenuation over fibre due to refractive index, its just less than the additive resistance of copper. A single optical fibre attenuates about 3dB over 15kms. Anyone know the attenuation of this fibre? Is this available in holey fibre? Having to run an amplifier every 15kms over the Atlantic would be expensive. Holey can go up to around 100kms, but is expensive.

Join the discussion!

Trending Stories Right Now