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

Re: Will HTTP/2.0 be green ?

From: Roberto Peon <grmocg@gmail.com>
Date: Sun, 1 Jun 2014 13:31:40 -0700
Message-ID: <CAP+FsNdv4V8JnkYnfNMg7m_ju6NtZu4ip0zW15yFCOXqTdMbMw@mail.gmail.com>
To: Poul-Henning Kamp <phk@phk.freebsd.dk>
Cc: HTTP Working Group <ietf-http-wg@w3.org>
So, lets see.. last time we looked, we were seeing something like a 15%
reduction in latency as measured by PLT to many of the most frequented
sites on the 'net.
The time to page usefulness, which is far more difficult to measure, was
improved more, meaning that the page could be navigated to/away from faster.

The trivial tests are most often the ones that get you the most meaningless

On Sun, Jun 1, 2014 at 1:27 PM, Poul-Henning Kamp <phk@phk.freebsd.dk>

> In message <CAP+FsNehJkfSsnJ9=
> 1d6joCgsskuyfiiUGb5fGp4L0QYgVO7SQ@mail.gmail.com>
> , Roberto Peon writes:
> >You're missing out on all of the other second order effects.
> >[...]
> >CPU is often far less costly than keeping the radio or screen on.
> You're welcome to count any and all secondary effects of HTTP/2.0
> to its advantage, provided you can show they exist, or even better
> if you can actually measure them.
> It would be pretty trivial to set up two Soekris boxes with a
> WLAN connection and measure the total power consumption running
> the same HTTP workload through HTTP/1.1 or HTTP/2.0.
> Poul-Henning
> --
> 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 Sunday, 1 June 2014 20:32:07 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:14:31 UTC