Re: Performance implications of Bundling and Minification on HTTP/1.1 from Poul-Henning Kamp on 2012-06-22 (ietf-http-wg@w3.org from April to June 2012)

From: Poul-Henning Kamp <phk@phk.freebsd.dk>
Date: Fri, 22 Jun 2012 22:05:51 +0000
To: Henrik Frystyk Nielsen <henrikn@microsoft.com>
cc: William Chan (???) <willchan@chromium.org>, HTTP Working Group <ietf-http-wg@w3.org>, Howard Dierking <howard@microsoft.com>
Message-ID: <2423.1340402751@critter.freebsd.dk>

I was referring to earlier discussions where an initial window of
six requests per TCP connection was proposed, based on the current
browsers "6 TCP connections" default param.

It is much easier and painless for a server to ignore SYN packets
than to tell every client "Yeah, I know the RFC says you can send
me 6 requests in parallel but..." for every SYN accepted.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.

Received on Friday, 22 June 2012 22:06:15 UTC