Japan Achieves 402 TB/s Data Rate - Using Current Fiber Technology (tomshardware.com) 21
Tom's Hardware reports that Japan's National Institute of Information and Communications Technology (working with the Aston Institute of Photonic Technologies and Nokia Bell) set a 402 terabits per second data transfer record — over commercially available optical fiber cables.
The NICT and its partners were able to transmit signals through 1,505 channels over 50 km (about 31 miles) of optic fiber cable for this experiment. It used six types of amplifiers and an optical gain equalizer that taps into the unused 37 THz bandwidth to enable the 402 Tb/s transfer speed. One of the amplifiers this was demonstrated with is a thulium-based doped fiber amplifier, which uses C-band or C+L band systems. Additionally, semiconductor optical amplifiers and Raman amplifiers were used, which achieved 256 Tb/s data rate through almost 20 THz. Other amplifiers were also used for this exercise which provided a cumulative bandwidth of 25 THz for up to 119 Tb/s data rate.
As a result, its maximum achievable result surpassed the previous data rate capacity by over 25 percent and increased transmission bandwidth by 35 percent.
"This is achievable with currently available technology used by internet service providers..." the article points out.
"With 'beyond 5G' potential speeds achievable through commercially available cables, it will likely further a new generation of internet services."
As a result, its maximum achievable result surpassed the previous data rate capacity by over 25 percent and increased transmission bandwidth by 35 percent.
"This is achievable with currently available technology used by internet service providers..." the article points out.
"With 'beyond 5G' potential speeds achievable through commercially available cables, it will likely further a new generation of internet services."
Slow progress.... (Score:2)
That's rather a slow progress from what we were used in technology:
https://tech.slashdot.org/stor... [slashdot.org]
At first I thought it was a dupe but I can't find anything else.
Re:Slow progress.... (Score:5, Interesting)
It's a big step up in the ability to produce high frequency amplifiers. It's cutting edge science.
I wonder if the other infrastructure is ready though. A quick search gives a value of about 1,300 terabits/sec for total internet traffic a couple of years ago, so presumably it's more now. Even so, this is a decent chunk of the entire internet's traffic down a single fibre.
I have visions of routers melting when DDOSed.
2023 record of well over 2PB/sec (Score:1)
In 2023, another group demonstrated 22.9 petabits/sec (almost 3 PB/sec) [nict.go.jp], which according to the linked article "demonstrated a record-breaking data-rate of 22.9 petabits per second using only a single optical fiber, which was more than double our previous world record of 10.66 petabits per second."
The recent 402TB/sec achievement was done over commercially available technology. The nearly-3PB/sec record from 2023 used research technology.
Records set using off-the-shelf technology are interesting even if th
Re: (Score:2)
Dang that's fast Re:Slow progress.... (Score:1)
300bps should be enough for anybody.
Re: Dang that's fast Re:Slow progress.... (Score:2)
My typing speed exceeded the minitel V23 modem's upstream channel at 75 bps.
Wireless over fiber for short-range (Score:1)
20 years from now everyone might be using fiber instead of copper pairs, even for short runs in the home
I see wireless dominating both fiber and copper except in cases where wireless isn't up to the task, isn't cost-effective, or where the data channel is "fixed" for a long period of time (years).
I see copper dominating cases where power-delivery is an issue (PoE or tomorrow's equivalent).
As far as fiber in the home and home-sized places in the coming decades: It will be an option but wireless will rule th
Re: Wireless over fiber for short-range (Score:2)
I don't see that ever being the case. Wifi delivers 400 Mbps across floors at best in my home. 10gig ethernet delivers 25x as much, including power.
And of course, there are faster versions of ethernet.
We need better protocols for wireless. (Score:2)
We need better protocols for wireless. And we need them yesterday!
I agree wireless SHOULD be the optimal solution (no need to connect things or hide cables). But we seem to make no progress beyond the fake numbers on boxes going up. And wireless companies getting money every year to churn out something 'new' they won't support for long.
That's where copper excels. You plug it in and you're likely done, or it's very obvious what to try next (replace cable).
I've never tried to install fiber so I don't know
Size of a 402 TB/s 50km data mule? (Score:1)
How big of a convoy of truck-mounted data centers would you need to make a 402 TB/s 50km data mule [wikipedia.org]*, assuming a 30-minute latency (that is, travel time)?
I would do the math but I'm not up-to-date on the highest-density commercially available storage media.
* For those whose first response to this post is "challenge accepted," assume that each truck is carrying a shipping-container-sized always-on, self-powered (generator + fuel or fully-charged batteries)** data center that is ready to be plugged into a high
Starlink (Score:1)
Starlink is the only viable way to connect rural people to broadband in either Japan or the USA -- it doesn't matter. Put all communications research on hold and subsidize Starlink instead for about a year or two, then resume funding research. I don't like Elon Musk (he stole the idea of Starlink from WorldVu when they went to him for launch services), but solving rural broadband is more important than hating someone. Logically, Starlink or the is the only way forward for humanity.
Re: Starlink (Score:2)
Fix Your Headline... (Score:4, Insightful)
"TB/s" is terabytes per second. "Tb/s" is terabits per second. The headline should read "Japan Achieves 402 Tb/s Data Rate".
Re: Fix Your Headline... (Score:2)
Came hear to say that. Can't believe slashdot would make this mistake.
Re: (Score:1)
Can't believe slashdot would make this mistake.
You must be new here ...
I'd rather commoditize than push the top end. (Score:2)
Meaning cheaper, easier, more reliable, etc.
I realize service providers have incentives to group lots of customers into fewer, faster lines... But I'd rather have peer to peer between houses with multiple connections to route things through. Heck throw in a few longer distance wireless nodes too (I forget the name of the peer 2 peer product I saw for mesh networks). I can see like 15 networks from my house, and I don't have THAT many close by neighbors (suburbs).
I get tired of the shenanigans from my 2 o
aaah, someone needs to learn the difference (Score:1)