From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.vyatta.com (mail.vyatta.com [76.74.103.46]) by huchra.bufferbloat.net (Postfix) with ESMTP id BB0682004FD for ; Wed, 8 Jun 2011 10:30:13 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by mail.vyatta.com (Postfix) with ESMTP id 511BD182923B; Wed, 8 Jun 2011 10:50:47 -0700 (PDT) X-Virus-Scanned: amavisd-new at tahiti.vyatta.com Received: from mail.vyatta.com ([127.0.0.1]) by localhost (mail.vyatta.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DYIW6fBYnrLz; Wed, 8 Jun 2011 10:50:46 -0700 (PDT) Received: from nehalam.ftrdhcpuser.net (static-50-53-80-93.bvtn.or.frontiernet.net [50.53.80.93]) by mail.vyatta.com (Postfix) with ESMTPSA id 158B9182911D; Wed, 8 Jun 2011 10:50:46 -0700 (PDT) Date: Wed, 8 Jun 2011 10:50:44 -0700 From: Stephen Hemminger To: Dave Taht Message-ID: <20110608105044.62a912af@nehalam.ftrdhcpuser.net> In-Reply-To: References: <1307537773.3057.17.camel@edumazet-laptop> <1307545032.3057.45.camel@edumazet-laptop> <1307546871.20906.223.camel@probook> <1307547908.3057.53.camel@edumazet-laptop> <1307551300.3057.71.camel@edumazet-laptop> Organization: Vyatta X-Mailer: Claws Mail 3.7.9 (GTK+ 2.24.4; x86_64-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Cc: Thomas Graf , jdb@comx.dk, bloat Subject: Re: [Bloat] Notes about hacking on AQMs X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Jun 2011 17:30:13 -0000 On Wed, 8 Jun 2011 10:52:07 -0600 Dave Taht wrote: > On Wed, Jun 8, 2011 at 10:41 AM, Eric Dumazet wr= ote: > > Le mercredi 08 juin 2011 =C3=A0 09:51 -0600, Dave Taht a =C3=A9crit : > > > >> > >> And they are *all* wrong to varying extents, which is why I like the > >> 'mondo classifier' idea for DSCP+firewalling mentioned earlier on this > >> thread. Converging on several standards for packet marking vs the > >> adhoc-ness of thousands of different partial solutions that now exist > >> really makes sense to me. > > > > I can tell you there are hundred of different *valid* setups, especially > > in server farms, when you want some control of network trafic, now > > machines have Gb or 10Gb links... >=20 > Well, there are hundreds of thousands of completely ad-hoc solutions > of varying degrees of effacy. >=20 > Getting it down to mere hundreds would be be a good start. >=20 > > Really, there is no "one big thing that solves all problems, > > automatically". >=20 > Oh, I agree. It isn't just a Linux problem. Cisco and Juniper have been doing QoS solutions for years. Like Linux there is the "billions of knobs version" and the KISS version. The KISS versions are fair queueing based. The problem is that the more complex QoS variants can't be done in ASIC's and go down the software path. Linux has the same problem, the more complex QoS ends up requiring locks that embed performance.