Is that a FAT PIPE or are you just pleased to stream me? TERABIT fibre tested Proximus and Huawei have successfully trialled a super-channel optical signal, flinging out information at up to one terabit per second (Tbps). Tech lothario Huawei shacked up with Belgian box-wrecker Proximus back in January. The pairing has now produced a single super-channel optical transport network (OTN) card with a …
I don't care if this turns out to be an unpopular opinion, but OMG! reg, please fix your headlines. I enjoy the occasional immature headline or something with a story/tradition/inside joke, but we are way way past that point. These silly titles are cringe-worthy, clichéd, done to death and rarely ever make sense from a tech stand point, and even more rarely give you an idea about the actual story.
Please, I enjoy the site, the content, the editors, the comments, the adverts, the anonymity when making this comment and later agreeing with myself, everything!!! but the damn headlines. They are utterly unclever and stupid.
Surely the units boil down to bits per cycle. My interpretation being that as 1 Hertz means 1 cycle per second:
5.7 bits per second per Hertz
=
5.7 (bits/sec)/(cycles/sec)
=
5.7 bits/cycle
Essentially they're talking about the bit-rate, versus the light-wave frequency (or cycle-rate) so the 2 rates per second cancel each other out leaving only the bits and the cycles. If you used a higher frequency carrier signal, you'd get a higher bit-rate and the impressive thing is increasing the bit-rate without needing to increase the carrier signal.
Please feel free to correct me if I'm wrong.
"Proximus/Huawei's transmission speed was conducted over a 1,040km fiber link using an advanced "Flexgrid" infrastructure with Huawei's Optical Switch Node OSN 9800 platform."
There's nothing new about Flexgrid - it's an ITU standard and some of ALu's, and probably other vendors, optical products already support it.
"And who cares about getting terabit rates to the last mile when consumers don't really need more than gigabit which can be done easily with copper?"
Firstly: Let me know when you have symmetric Gb/s on _existing_(*) copper from the local phone cabinet to the customer's premises without using some crappy power sucking device that requires perfect lines less than 100 feet long and breaks as soon as you breath hard on it.
Secondly: Having gotten all those GB circuits to the cabinet, at what speed do you intend to backhaul them to the local central office/concentrator node? What speed do you intend to backhaul from the concentrator/CO to the regional distribution node?
(*) if you have to run new lines then they may as well be glass.
You can upgrade FTTN just as easily, can't you?
Only as far as the node. FTTN leaves non-upgradeable copper for part of the system, which makes the *entire* system throttled to the slow speed of the non-upgradeable part.
consumers don't really need more than gigabit which can be done easily with copper?
Remember the classic quote "640 K ought to be enough for anybody"? Didn't turn out so well, did it!
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
