Re: Comparing HTTP over QUIC header compression schemes

[Patrick McManus <[e-mail]>]

Alan, I think this is really interesting. I want to reserve comment until I
can read it when not multipexing with a WG meeting :)

on the PLT: fastly generously shared some of their US based loss rates here
https://www.slideshare.net/Fastly/http2-what-no-one-is-telling-you (slide
102) .. which shows > 1.5% on roughly ~12% of connections. There are a
million caveats of course - but its one of the broader datapoints I've seen
and its awesome they shared operational information.

On Tue, Jun 6, 2017 at 11:10 AM, Martin Thomson <[e-mail]>
wrote:

> On 6 June 2017 at 10:59, Eggert, Lars <[e-mail]> wrote:
> > (2) simulating loss rates much above 1% is likely going to be pretty
> uninteresting, given that QUIC (and TCP, FWIW) have congestion controllers
> that will struggle to even deliver useful throughputs in these cases
>
>
> That's an interesting point.  I've seen a presentation (that I can't
> find now, but I can try) that shows that packet loss in real-world
> scenarios approaches 2% in a great many cases.
>
>