Microsoft and Facebook's transatlantic cable completed Construction of Microsoft and Facebook's jointly-funded submarine cable has ended. The two companies, along with construction and operation partner Telxius, revealed the project in May 2016 and promised a 160 terabit per second link across a 6,600 km route linking Virginia Beach, USA, and Bilbao, Spain. Cable-laying work …
It will be interesting to see just how much transatlantic capacity will bypass UK in the coming years.
This is the first fat fiber to do so in its entirety. While there are a couple that go around the UK to the north of it and land Denmark as well as a couple going France but they are pretty old (and full) - from the dot bomb age. They also have UK involvement (the protection circuits go via the UK).
Leaving politics aside, this is good for the Internet resilience. All it takes to massively cripple it today is one plough towed right outside UK territorial waters around Cornwall.
Sennen Cove.
Beware militant surfers? Sea Lions can be dangerous I guess. I seem to recall the first transatlantic cable landed in Sennen Cove did it not?
Related geeky interest, whilst in that neck of the woods check out the telegraph museum: http://telegraphmuseum.org/
Interesting enough for a visit even on a dry day!
@ Steve Davies 3 ........ "I can just imagine a farmer on his tractor and ploughing the seabed off of Sennen Cove"
Well the sea bed off Sennen Cove and Porthcurno too if you want to be really effective. Alternatively if you want a more suitable place for a farmer on a tractor you could just go inland a bit to Skewjack farm and plough everything up there ..... but if you do, expect GCHQ to be miffed. They wont like you making holes around their interception station.
It will be interesting to see just how much transatlantic capacity will bypass UK in the coming years.
I see you're trying to make some sort of point. Cable is only cost-effective because of the Bilbao-UK link but hey. Real reason it's there because people are starting to realise that running everything through New York is a flaky plan; most of us knew it about 2 decades ago, only thing that surprises me it took so long to bring online.
What's the good of bypassing the UK when the NSA is a thing. "Avoids five eyes" - find grip. Know what else is in Virginia? The Pentagon, the CIA HQ who knows how many NSA sites. Avoids five eyes; Virginia is five eyes central.
Question:
If you lay a transatlantic cable, and then over the intervening years ten other people come and do the same, it's virtually inevitable that they will cross over.
When they do, and there's a problem, requiring someone to get a big boat and go out and pick up your cable, how do they avoid tampering with all the others? Surely they must end up dragging those up too, and then how do you fix and relay your own tangled cable and (if you break it) theirs without repeating the situation?
I realise that most of the time there's enough slack to pick the cable up anyway (or you wouldn't be able to repair) but surely it must be an issue?
Not being nautically-minded myself, I hadn't thought of conjoined anchors.
But it just seems to me that such cables would be very vulnerable in such instances. I know the ships can splice and repair, but it just seems a problem that can only get worse, and even when it comes to "remove that 50-year-old obsolete cable", you will still have the same problem.
And what happens in, say, a war, and you can't survery 6000km of cable, and you can claim "Whoops, sorry, we were just repairing OUR cable but happened to take out your continent's fibre in the process". They must be more valuable than oil pipelines nowadays, no?
but surely it must be an issue?
I would guess that it is only an issue in relatively shallow water at hub locations where multiple cables come ashore together. Out in the really deep water, I'd expect cable layers plan to avoid other vaguely parallel cables by general routing (given that most have low accuracy position data). That's also what it looks like in the submarine cable map that colinb posted the link to. Obviously there's situations where you have to cross another cable, I think all they can do there is cross their fingers and hope the probability of getting a fault close to the crossing point is probably very low, in the entire 6,000 kms plus of cable.
If there is, then the cable repair ship has to go trawling for the cable, and hope that they pull up the correct one (its a very inexact science).
In the late 90s I went to a networking conference at which the skipper of a C&W (I think) cable laying ship gave us a very entertaining lecture about the various superlatives, stresses and challenges involved in putting fibres in the ocean.
Repairing broken cables involved dragging miles of cable to the surface, slicing it into two pieces and looking which side the light came from, then repeating this on the broken side until light was found again, then splicing in a new section.
He ended with a tale of being tasked to locate and repair a cable near the Canary Islands.This was something like 3 miles under water, so they put down a grapple and towed across where the cable should be. After snagging something, they lifted it and committed the slice... only to find light on both sides.
Oops. Wrong one.
They were called rather rapidly with a report of another cable failure, which they graciously fixed before doing the one they were supposed to do.
Not an exact science as someone else said!
"Repairing broken cables involved dragging miles of cable to the surface, slicing it into two pieces and looking which side the light came from, then repeating this on the broken side until light was found again, then splicing in a new section."
You'd think they have some sort of management system or at least an OTDR to help them find that place faster. After all, every cut means additional attenuation which means more noise and therefore, in the long run, more problems with more advanced modulation schemes.
160 terabit per second
Now thats speedy!
I also like how unambiguous it is about wether its bits or bytes. Refreshing.
When ISPs quote things like "ooh u can get 50 mbs" or mbps or Mb/s Its never clear wether they are indicating bits or bytes , or if they even know they know the difference , I guess the techs do , do the salesdroids? Are they pulling the wool over our eyes by relying on the customer not knowing and quoting the bits implying they are bytes?
I seem to remember the correct distinction is all in the capitals or something?
How does one correctly quote bits and bytes?
"My understanding is that Bytes are capitals and bits are lowercase, and that all the multiplications are capitalised"
Close. All prefixes greater than 1000 use a capital letter. Kilo, hecta and deca use lower case (the latter being the only SI prefix not to use a single letter), while all multiplications smaller than 1 use lower case (although obviously these aren't used much in conjunction with bits and bytes).
In telecoms it's always bits.
There are several reasons, but one is that you can't be certain that 8 consecutive bits belong together to make a byte, or that the bits will be assembled into bytes at all - the concept of a byte of data transmitted across a system exists at a different layer to the transmission mechanisms of that system.
In computing 'B' is bytes, 'b' is bits.
But neither MS or FB are short of a few $$$ and this cable is all tax deductable (not that either of them pay tax on their profits outside the USA in the first place).
Eight fibre pairs is very small but this article seems to explain why
http://www.independent.co.uk/life-style/gadgets-and-tech/everything-you-need-to-know-about-the-undersea-cables-that-power-your-internet-and-why-theyre-at-a6710581.html
IMHO, the biggest cost is to hire the cable layer not the cable itself.
"How many cat videos per second will fit down the inner-tube?"
I've just checked The Register Standard units page and there doesn't appear to be a measurement unit for data transfer. Which is surprising! I think CV/Ps should be considered for entry. You'd need the average length of a cat video though.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
