I concur with Derek's logic. Also, I do networking for an ISP.
It's the peering/transit point that is at capacity and there's a "layer
8" (politicial/$$) issue stopping the upgrade.
There's many analogies that can be made (and are being made), but it
basically comes down to the fact that the more that a single entity uses
your network, the more you become beholden to them and their
demands/needs, and that's a huge risk - doubly so when they're not
paying you. (That's actually something to be said about business in
general).
So it's a power struggle and you're stuck on the sidelines watching the
fight. Your only power is to change providers and hope enough people do
it as well. You can partially do that by using a vpn proxy service.
As a side, it's generally not economically worth doing individual (tcp)
stream bandwidth management at the ISP scale - (we're talking billions
of streams at this point). Also, we don't care about the contents of
your streams - just the source/destination - the NSA/CIA probably does
though. Bittorrent taught us that lesson.
In fact I've been seeing more companies start worrying about their staff
doing exactly what Edward is doing (vpn relay through the office) and
potentially having to pay for that extra bandwidth in/out. That's where
bandwidth management becomes economically viable.
R.
On 2014-07-22 10:01 AM, Derek Balling wrote:
So I VPN'd into work (We have a non-split-tunnel VPN available), and
then we can watch it, no problem. It's the same content, being
delivered over the same network, only it's encrypted and hidden from
FiOS's routers. There's no other explanation, simply, caught red handed.
I reject the premise.
By VPN'ing into a different network you are -- by definition -- getting
traffic to and from Netflix via a different peering-point, presumably
one that doesn't have 100s, 1000s, tens of 1000s, of Netflix customers
all trying to cross it at the same time. It may be a narrower peering
point than the one FIOS is connected to, but it suffers FAR less contention.
You've constructed a flawed experiment, and then extrapolated massively
from it.
D
_______________________________________________
Discuss mailing list
[email protected]
https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss
This list provided by the League of Professional System Administrators
http://lopsa.org/
_______________________________________________
Discuss mailing list
[email protected]
https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss
This list provided by the League of Professional System Administrators
http://lopsa.org/
