In message <[email protected]>, at 21:29:44
on Sat, 22 Mar 2014, Gordon Lennox <[email protected]> writes
"For most of us, the internet is what you’re looking at right now—what you
see on your web browser.
Which is fine for content on a web page, but you and I are looking at
emails "right now".
But the internet itself is comprised of the fiber optic cables, the
servers, the proverbial series of tubes, all owned by the companies
that built it. The content we access online is stored on servers and
transmitted through networks owned by lots of different groups, but the
magic of the internet protocol lets it all function as the integrated
experience we know and, from time to time, love."
http://qz.com/187034/how-the-internet-works-and-why-its-impossible-to-know-what-makes-your-netflix-slow/
Well maybe sort of...
It's more of a "where might the bottlenecks be", and focusses too much
on a US-Centric view of national connectivity architecture. And the
whole idea of "Tier 1" and "Tier 2" is so 1990's.
One part they leave out is contention (or under-capacity) *inside* the
ISP at the end of the 'last mile'. Here in the UK it's that bit of the
network which matters too.
My connectivity has a 'last mile' (unusually it is about a mile, for
many subscribers it's more) over legacy copper, and I can pay to have
better technology applied to squeeze more data down it; but then there's
a '15 miles' to a regional hub and a '50 miles' to London, both of which
are still *inside* my ISP and both of which affect the speed I observe
[anecdotally it's how much the ISP spends on provisioning that '15
miles' which is most crucial].
These three bottlenecks (1, 15, 50 miles) are an extra layer that
affects all content of course, before we start wondering if there's a
priority between that router in London and servers run by YouTube,
Netflix, or BBC iPlayer.
--
Roland Perry
