Knock out this set of buildings in London, and you can kiss the entirety of the UK offline/most of Europe won't recover: https://en.wikipedia.org/wiki/Telehouse_Europe

It took exactly one (1) spanning tree misconfiguration on a single BlackDiamond to cripple LINX for hours, a couple of decades ago.

Source: I was sitting next to the person who fixed it.

You’re wrong. If say One Wilshire[1] or one of the very few other carrier hotels in its class abruptly ceased existing the Internet would be wrecked and rebuilding would be a matter of months at the very best. It doesn’t matter how many peers a big telco has when the supermajority of backbone peerings are in a handful of buildings.

[1] https://one-wilshire.com/

I used to have a security badge to get in there back in 2003 or so. When I was shown around someone said they could cause a minor recession in 5 minutes with an axe. Also they had me grab a fibre that carried all of Japan’s internet traffic. If you pulled that it would route everything the other way around the world through Europe and the east coast.

I had a long lost picture when they were doing some street construction outside and marked all the buried cables with spray paint. Lines going everywhere…

First off, I'm on the east coast-ish, so I'm not sure how much traffic I get from APAC.

Next off, the their front page says:

> Carrier-Neutral with Access to 200+ Carriers

You're telling me these 200+ carriers have no other POPs?

Going to their "Connectivity" page, they show four trans-Pacific cable that land there:

* https://one-wilshire.com/connectivity/

Meanwhile there are a whole bunch of other trans-Pacific cables landing on the US West Coast:

* https://www.submarinecablemap.com

As someone who lives in Toronto, I certainly worry about how concentrated TORIX is, but even it has three physical locations (IIRC), and most folks connect elsewhere as well from the spelunking I've done in BGP.

So if one of these major carrier hotels does have a problem, I don't doubt that there would be repercussions, but I'm not worried about "the Internet" as a whole.

...maybe. Pretty sure the "wreck" would be local to the region around LA, and maybe a few of the countries on the other end of those lines. The immediate effect of an unexpected outage (e.g. the physical destruction of the building) would probably be a lot of rerouted traffic overwhelming systems "nearby" (in the topology of the Internet) for a few hours. It would not take months to recover even in the absolute worst case.

As a network person who's worked at multiple major global transit networks: lol no.

That building gives me anxiety attacks; just thinking about how vulnerable it is.

Not the whole internet, but a whole country for sure: google "Las Toninas" a beachside town in Argentina. Someone over there frequently steps over the wire and good bye rest of the world

I didn't RTFM but I could see the argument that the SPOFs of the Internet are AWS, Cloudflare and Google, not the cables and routers in an IXP.

You alread listed 3 single points of failure. I get what you mean: the entry point search engine, the global cache and the global backend/storage/compute.

But look I live in Hong Kong, I dont feel that way: there are other backends, we could survive without the caching for a while and Google is forbidden for 1.4bn people who get on with it very well...

Depends what you call the internet. Yes, facebook and whatsapp are gone the minute one of those 3 companies screwed up.

OK, you think 1.4 billion people without the freedom to search is ok? LOL.

I wouldn't say Google is search, per se. I haven't used Google for search in a few years.

If Google search went away for more than 24 hours, Bing/DDG/others would quickly get overwhelmed as people learned about alternatives and switched.

Bing might be able to auto-scale using Azure infrastructure but there might not be enough hardware even there.

WWW != Internet

Yes, for the majority of users the two may be the same, but "The Internet" is the IPv4 and IPv6 network on which the web runs. The equivalent companies to what you named are the transit network operators, companies like Hurricane Electric, AT&T, CenturyLink, etc. The list of transit networks is very long, much longer than the list of major web hosting companies.

Now, that said, it is absolutely true that there are ways where major outages can be the result of just one failure. Last year there was a major outage on the US east cost after a Verizon fiber optic line was severed in NYC; it turns out that this line carried many supposedly redundant links, all bundled into a single cable. The failure of that line then caused a cascade of failures as rerouted traffic began overwhelming other systems, but in the end the outage was contained in a relatively small geographic region, albeit one with many Internet users.

The article is about BGP and Internet Routing Registry (IRR): routing.

Those are not single points.