From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lb0-x235.google.com (mail-lb0-x235.google.com [IPv6:2a00:1450:4010:c04::235]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 79E7221F3E7 for ; Tue, 6 Jan 2015 13:49:24 -0800 (PST) Received: by mail-lb0-f181.google.com with SMTP id l4so81037lbv.40 for ; Tue, 06 Jan 2015 13:49:22 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=j5P2SF2lv6A8OuIageWyZ0TMBgrpFCKgeekk+kQm15g=; b=un4wSXEfizX5vWmxDZ5vYS0CKMknPy26fszoblZdw6HxUxwAJBLxgxjXKwlpikTJak gF7iPp8DoM6YY+NM+1sTOZv2IMwEGqAVfDUXOHFc52Nwed6zIczVk67tWrLtgVU1tlYn 06MRuafqEAc/Kkmr2acuK0X9Z2lu6PmjLSYffHIZ5A43b5PIDyg27nTEunL/wKToGalp UHCp/9KUgrbcyZCCRfolXS84EnO3VfWdwtt5es987xENJ9KpBBKVxghTgrPtIIeBag33 Ak1dU/Fh4ETc6DSLPTO/Y/JgEc9bSiY6SbyqJSZJGnj4/pvdouGVLs6DzY66+rsE3tGu bedA== X-Received: by 10.112.130.34 with SMTP id ob2mr43383309lbb.78.1420580961797; Tue, 06 Jan 2015 13:49:21 -0800 (PST) Received: from bass.home.chromatix.fi (178-55-208-88.bb.dnainternet.fi. [178.55.208.88]) by mx.google.com with ESMTPSA id zs7sm15635357lbb.18.2015.01.06.13.49.20 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 06 Jan 2015 13:49:20 -0800 (PST) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 8.1 \(1993\)) From: Jonathan Morton In-Reply-To: <20150106204528.8FAE240600C@ip-64-139-1-69.sjc.megapath.net> Date: Tue, 6 Jan 2015 23:49:18 +0200 Content-Transfer-Encoding: quoted-printable Message-Id: References: <20150106204528.8FAE240600C@ip-64-139-1-69.sjc.megapath.net> To: Hal Murray X-Mailer: Apple Mail (2.1993) Cc: bloat@lists.bufferbloat.net Subject: Re: [Bloat] Bufferbloat: GoGo blocking YouTube 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: Tue, 06 Jan 2015 21:49:53 -0000 > On 6 Jan, 2015, at 22:45, Hal Murray wrote: >=20 > GoGo provides internet access on airplanes. They want to block = YouTube=20 > (and similar) to avoid overloading their thin pipes. They are doing = that by=20 > intercepting https connections and presenting bogus certificates. =E2=80=A6WTF? Look - if you *want* to block YouTube, then you block YouTube. People = might get a little annoyed about that, but it=E2=80=99ll probably be = limited to minor grumbling. You *don=E2=80=99t* fiddle with traffic to = it. There is something *seriously* wrong if that=E2=80=99s the first or = best solution that came to mind. I agree with the conclusion of the article, though. There=E2=80=99s a = straightforward, network-neutral, technological solution which actually = solves the original problem. Shame almost nobody=E2=80=99s heard of it. Incidentally, I finally got my test setup running properly. It now has = cake running on each of two Fast Ethernet interfaces in the Pentium-MMX, = which are bridged. It is able to comfortably pass 50Mbps through that = before it runs out of CPU grunt - but that=E2=80=99s 50Mbps total. It = doesn=E2=80=99t matter whether it=E2=80=99s all one way, all the other = way, or half each. I then set it up to simulate a 24/3 Mbps ADSL, and = it did that with about 50% CPU time in soft-interrupt mode. I haven=E2=80=99t tried cake2 yet. The limiting factor may well be context switching, or at least interrupt = handling overhead. That=E2=80=99s quite expensive on x86 and on a full = OS like Linux; far more so than on, say, an ARM running in a dedicated = embedded configuration. (ARM has banks of registers which are switched = in, replacing the originals, for interrupt handlers, so it doesn=E2=80=99t= have to hurriedly save all those registers before it can do anything = useful.) Bridging, and running *both* the traffic endpoints on other = machines, rather than keeping one endpoint on the Pentium-MMX, improves = the throughput markedly. - Jonathan Morton