W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2014

RE: HTTP/2 Priorities Proposal

From: RUELLAN Herve <Herve.Ruellan@crf.canon.fr>
Date: Tue, 22 Apr 2014 16:05:43 +0000
To: Martin Thomson <martin.thomson@gmail.com>
CC: Jeff Pinner <jpinner@twitter.com>, HTTP Working Group <ietf-http-wg@w3.org>
Message-ID: <6C71876BDCCD01488E70A2399529D5E532D4652A@ADELE.crf.canon.fr>

> -----Original Message-----
> From: Martin Thomson [mailto:martin.thomson@gmail.com]
> Sent: vendredi 18 avril 2014 18:41
> To: RUELLAN Herve
> Cc: Jeff Pinner; HTTP Working Group
> Subject: Re: HTTP/2 Priorities Proposal
> On 18 April 2014 08:11, RUELLAN Herve <Herve.Ruellan@crf.canon.fr> wrote:
> > I however have one concern: priority groups are structures that the client
> knows will be persisted by the server. As such, a client changing the priority of
> a group is certain that the server will be able to apply it.
> There are no guarantees.  Groups can be discarded in the same way that
> priority information for an individual stream can be.  There might be
> a greater incentive to keep it for longer, but it's still expendable
> in case of memory pressure.  (I note here that memory pressure had
> better be extreme for you to want to save this tiny amount, though I
> can devise schemes where it could make sense.)

I agree that there are no guarantees, but I think that if a group is discarded by the server due to memory pressure, then that means that the server is relinquishing any pretension to do any prioritization based on client hints.
On the other hand, a server need to discard closed streams and if it is too eager to discard top level streams, the client lose some ability to re-prioritize a whole sub-tree of streams.

Received on Tuesday, 22 April 2014 16:06:18 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:14:30 UTC