From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp86.iad3a.emailsrvr.com (smtp86.iad3a.emailsrvr.com [173.203.187.86]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 3AF833B2A4 for ; Sat, 28 Sep 2024 17:07:13 -0400 (EDT) Received: from app31.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by smtp27.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id C359A23AFC; Sat, 28 Sep 2024 17:07:12 -0400 (EDT) Received: from deepplum.com (localhost.localdomain [127.0.0.1]) by app31.wa-webapps.iad3a (Postfix) with ESMTP id 956592221E; Sat, 28 Sep 2024 17:07:12 -0400 (EDT) Received: by apps.rackspace.com (Authenticated sender: dpreed@deepplum.com, from: dpreed@deepplum.com) with HTTP; Sat, 28 Sep 2024 17:07:12 -0400 (EDT) X-Auth-ID: dpreed@deepplum.com Date: Sat, 28 Sep 2024 17:07:12 -0400 (EDT) From: "David P. Reed" To: "David Lang" Cc: "Dave Taht" , "Cake List" , "bloat" MIME-Version: 1.0 Content-Type: text/plain;charset=UTF-8 Content-Transfer-Encoding: quoted-printable Importance: Normal X-Priority: 3 (Normal) X-Type: plain In-Reply-To: References: <1727471439.369527853@apps.rackspace.com> X-Client-IP: 209.6.168.128 Message-ID: <1727557632.49023402@apps.rackspace.com> X-Mailer: webmail/19.0.25-RC X-Classification-ID: fb0eda4f-f2df-45bd-b2d1-1e2ba650ff05-1-1 Subject: Re: [Cake] bbr vs all the aqms, cake winning... X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 28 Sep 2024 21:07:13 -0000 =0A=0AOn Friday, September 27, 2024 17:43, "David Lang" sai= d:=0A=0A> =0A> could the community try and produce 'traffic simulators' tha= t implement these=0A> various protocols with a more realistic traffic patte= rn? something that can be=0A> turned up or down with a few presets of the m= ix that we can make available for=0A> the academics to use for their testin= g?=0A=0AIt's a pretty good idea.=0A=0AHowever, some things I think about...= it's important to remember that just recording traffic streams and playing = them back filters out almost all of the "control" dynamics.=0AIn the real I= nternet. It's the control feedback that causes instability and bufferbloat = - everything is feedback system among a large set of endpoints - typically = hundreds of them at minimum. It's hard for those who think from a perspecti= ve of one "server", or from one "client", or between a "pair" to realize th= at networks aren't like that. Every "independent" but concurrent activity f= eeds back significant "noise" into the other activities, changing latency i= n every protocol, etc.=0A=0AEven the dynamic behavior of a single router ou= t-queue affects everything, especially, as you know, the queues upstream an= d downstream.=0A=0ATHat's one reason that I find any work that tries to fin= d ways to store packets in router queues ("background service") to be dubio= us, because that's one of the things that always increases dynamic instabil= ity. Always is my belief based on much experience and experimentation - the= re may even be a theorem there, but it's tricky to find the premises of suc= h theorem.=0A=0AHowever, this suggests a "stress simulator" that might help= create realistic experiments - instead of simulating traffic, simulate ins= tabilities in routers by injecting packets into queues for mix of flows bas= ed on some probabilistic model that acts like "real Internet". These inject= ed packets might, for example, have invalid checksums on purpose to avoid h= arming any endpoint stacks. So such a tool would be a lot easier to deploy.= =0A=0ARRUL isn't typical, but it at least is a simple setup that pushes cer= tain AQM challenges into a real bottleneck queue, if there is one bottlenec= k. Simulating, say, a highly optimized "single page web site" programmed in= JavaScript with some rich UI packages, even in the single bottleneck case = with multiple edge smartphones running other apps (like Zoom and streaming = audio) - one can't really tell if that covers any other cases of interest.= =0A=0A> =0A> David Lang=0A> =0A