MUX and AJPv2 (was: HTTP-NG)

[Ian Kluft <ik...@cisco.com>]

> From: Daniel Veillard <ve...@rufus.w3.org>
> Ian Kluft <ik...@cisco.com> wrote:
> :It was enough to determine it was too much overhead for what we needed.
> :However, we made a note that the Apache Group would probably be interested
> :eventually.
> 
>   I would be interested in the analysis of Apache-Java Team w.r.t. the
> MUX protocol.

If I had somehow foreseen the flamefest about W3C that was about to erupt,
I might have elaborated.  Indeed, MUX was "too much overhead for what we
needed" but that was not considered a bad thing about MUX, just a technical
analysis of our project's needs.

We determined that the new Apache JServ Protocol needed to be able to
multiplex sessions on a single pipe in order to better support a
multithreaded client.  (The "client" from the point of view of this protocol
is the HTTPD.)  But when we considered the factors in implementing it on a
single-threaded client, we started considering restrictions on its use that
made it clear we didn't need as much generalization as MUX had to offer
(or to implement.)

We ended up accepting some influences from MUX and defined a protocol that
has multiplexed Apache JServ sessions on a single pipe.  It's now available
online in its interim form - we are currently doing implementation
experiments so the spec states it's still subject to revision:

   "Apache JServ Protocol Version 2 (AJPv2)"
   http://java.apache.org/doc/ajpv2-spec.html

Other project-related info can be found at http://java.apache.org/
-- 
Ian Kluft  KO6YQ PP-ASEL                                  Cisco Systems, Inc.
ikluft@cisco.com (work)  ikluft@thunder.sbay.org (home)          San Jose, CA