- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Fri, 27 Jun 2014 09:43:32 -0700
- To: David Krauss <potswa@gmail.com>
- Cc: Poul-Henning Kamp <phk@phk.freebsd.dk>, Mark Nottingham <mnot@mnot.net>, HTTP Working Group <ietf-http-wg@w3.org>, Roberto Peon <grmocg@gmail.com>
On 27 June 2014 03:01, David Krauss <potswa@gmail.com> wrote: > Perhaps I missed something, but who is implementing CONTINUATION? I only recall hearing from folks who were not. I don't know how many of these [1] implement CONTINUATION, but I'd imagine that it's close to 100%. I know that it makes some people sad, where perhaps a larger frame size would not. But of those that have HTTP/2 implementations, the number of complaints we've received is fairly low, if vocal. The more general issue of frame size and forwarding performance is probably less well-tested, probably because the focus thus far has been on functional interoperability. Performance on the scale that Willy and PHK concern themselves with takes a little more time and effort to reach. On the other hand, I hear that Google do test - and have tested - at scale and aren't concerned about performance, but it's not perfectly clear cut.
Received on Friday, 27 June 2014 16:44:00 UTC