- From: Cory Benfield <cory@lukasa.co.uk>
- Date: Wed, 30 Nov 2016 12:48:29 +0000
- To: Benedikt Christoph Wolters <benedikt.wolters@rwth-aachen.de>
- Cc: RFC Errata System <rfc-editor@rfc-editor.org>, Mike Belshe <mike@belshe.com>, fenix@google.com, Martin Thomson <martin.thomson@gmail.com>, ben@nostrum.com, alissa@cooperw.in, aamelnikov@fastmail.fm, Patrick McManus <pmcmanus@mozilla.com>, Mark Nottingham <mnot@mnot.net>, ietf-http-wg@w3.org
> On 30 Nov 2016, at 12:42, Cory Benfield <cory@lukasa.co.uk> wrote: > > >> On 30 Nov 2016, at 11:32, Benedikt Christoph Wolters <benedikt.wolters@rwth-aachen.de> wrote: >> >> 2016-11-30 9:41 GMT+01:00 Cory Benfield <cory@lukasa.co.uk>: >>> What happens if both stream A and B are blocked? Should my server endeavour to serve dependent streams in that case? I *think* the answer is yes because of the logic around having grouping nodes formed from idle streams, but I’m not 100% sure and would like clarification. >> >> Section 5.3.4 Prioritization State Management states: >>> Resources are shared between streams with the same parent stream, which means that if a stream in that set closes or becomes blocked, any spare capacity allocated to a stream is distributed to the immediate neighbors of the stream. >> >> In my understanding in your example A and B are blocked, C and D are >> immediate neighbors of A so according to this they would get the >> resources of A. > > No, they aren’t immediate neighbours, they are immediate *children*. This is exactly the core question. Oh, I apologise, I hadn’t realised that “neighbours” might be being applied in terms of its graph-theoretical definition. That’s an interesting point.
Received on Wednesday, 30 November 2016 12:49:04 UTC