S.Korea broadband firm sues Netflix after traffic surge

Owen DeLong owen at delong.com
Mon Oct 11 01:05:38 UTC 2021

> On Oct 10, 2021, at 13:21 , Mark Tinka <mark at tinka.africa> wrote:
> On 10/10/21 22:13, Michael Thomas wrote:
>> Isn't that what Erlang numbers are all about? My suspicion is that after about 100Mbs most people wouldn't notice the difference in most cases. My ISP is about 25Mbs on a good day (DSL) and it serves our needs fine and have never run into bandwidth constraints. Maybe if we were streaming 4k all of the time it might be different, but frankly the difference for 4k isn't all that big. It's sort of like phone screen resolution: at some point it just doesn't matter and becomes marketing hype.
> The ISP looking to charge BigContent for increased link saturation isn't looking at the individual 100Mbps links they have sold to their downstream customers.
> They are looking at the aggregate Gbps or Tbps of traffic that BigContent is seeking to deliver across their network, for "no $$".

Which is the kind of ignorant view of the situation that creates this problem in the first place.

It’s not for “no $$”, it’s for all the $$ they got from all those 100Mbps links that they are delivering those Tbps of traffic to.

If the aggregate $$ they are collecting is insufficient, then they have priced their service incorrectly and should either re-evaluate,
or go bankrupt and sell to someone that knows how to run a business.


More information about the NANOG mailing list