From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-f53.google.com (mail-wm0-f53.google.com [74.125.82.53]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 52B523B2A4 for ; Tue, 12 Jun 2018 10:30:09 -0400 (EDT) Received: by mail-wm0-f53.google.com with SMTP id p11-v6so23755715wmc.4 for ; Tue, 12 Jun 2018 07:30:09 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=NXQ60B4HSie/J7ZnuM17fwB05AGYUbd/0/lv484R3hY=; b=LXrKAUMSlFCZTUmCbzAKaDOtc+bj08NfPRiO2gezMkgxlod8Wu2nc3Ev0Usc8IfqBx 5cwueCF6msxv3Ms1Ci8jOJKZYKaaaI8GHrH4Hve515OwL2Qa6+VOlHlJ8ZYaxcKN9OHN tXrJy+01uatcI6LKC0sr2EcTr+Ugrt4MXWuZkrIbsTvVFFsvmhP7n0qztQLT51TaaZfG h6Ezt6UhNO/hgvWusPlgkUHwmKuLbpdqanlN+JYf8Yh3ey3TixFKNm24yMQLdbVBMBPU 3I6i5bZUsWkxfPzyOUp5mLlrpHVPBm81y3C+CGA5/1pbQFyXdCProkq5zmZVNUWHwQR5 QUjg== X-Gm-Message-State: APt69E3YG0ftTmKfOYwdf9chfp5Ug5QCFBtl7rSVVyyIUoWw45TsknEL cCQOfsxpgjzUyrcet/nnxuOj0GqQrLm9cN8AvEg= X-Google-Smtp-Source: ADUXVKKFyJuqknrsKBrSdJ/JJHFI2i749xg35QpdeK2DqzyKDsM70GDRCc+SRqketQZsUWXHHfzKvXiswhEfGr+03uk= X-Received: by 2002:a50:9298:: with SMTP id k24-v6mr614523eda.110.1528813808240; Tue, 12 Jun 2018 07:30:08 -0700 (PDT) MIME-Version: 1.0 References: In-Reply-To: From: Jim Gettys Date: Tue, 12 Jun 2018 10:29:50 -0400 Message-ID: To: Dave Taht Cc: kevin@darbyshire-bryant.me.uk, gih@apnic.net, bloat Content-Type: multipart/alternative; boundary="000000000000f7cd89056e72b361" Subject: Re: [Bloat] geoff huston's take on BBR X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Jun 2018 14:30:09 -0000 --000000000000f7cd89056e72b361 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, Jun 12, 2018 at 2:55 AM Dave Taht wrote: > On Mon, Jun 11, 2018 at 10:58 PM, Kevin Darbyshire-Bryant > wrote: > > > >> On 11 Jun 2018, at 22:27, Dave Taht wrote: > >> > >> https://ripe76.ripe.net/presentations/10-2018-05-15-bbr.pdf > > > > Fascinating! > > > > " =E2=80=A2 BBR changes all those assumptions, and could potentia= lly push > many networks into sustained instability > > > > =E2=80=A2 =E2=80=93 We cannot use the conventional network con= trol mechanisms > to regulate BBR flows > > > > =E2=80=A2 Selective packet drop just won=E2=80=99t create back pressure= on the flow=E2=80=9D > > > > And I keep on seeing questions on whether BBR understands ECN - if not= =E2=80=A6. > well I think we see the results. > > I think geoff goofed in his testing of BBR, starting all flows at the > same time, thus syncing up their probing periods. Real traffic is not > correlated this way. > (I made the same mistake on my initial bbr testing) > =E2=80=8BBeware of self synchronization... In other situations, there have= been situations in which aggregate behavior occurs due to this phenomena. So both synchronized and unsynchronized startup of flows under test is actually a good idea (and watching to see if anything in the aggregate behavior over time results in such synchronization). - Jim --000000000000f7cd89056e72b361 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

On Tue, Jun 12, 2018= at 2:55 AM Dave Taht <dave.taht@= gmail.com> wrote:
On Mon, = Jun 11, 2018 at 10:58 PM, Kevin Darbyshire-Bryant
<kevi= n@darbyshire-bryant.me.uk> wrote:
>
>> On 11 Jun 2018, at 22:27, Dave Taht <dave.taht@gmail.com> wrote: >>
>> https://ripe76.ripe.net/presenta= tions/10-2018-05-15-bbr.pdf
>
> Fascinating!
>
> "=C2=A0 =C2=A0 =C2=A0 =C2=A0=E2=80=A2 BBR changes all those assum= ptions, and could potentially push many networks into sustained instability=
>
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0=E2=80=A2 =E2=80=93=C2=A0 We cannot u= se the conventional network control mechanisms to regulate BBR flows
>
> =E2=80=A2 Selective packet drop just won=E2=80=99t create back pressur= e on the flow=E2=80=9D
>
> And I keep on seeing questions on whether BBR understands ECN - if not= =E2=80=A6. well I think we see the results.

I think geoff goofed in his testing of BBR, starting all flows at the
same time, thus syncing up their probing periods. Real traffic is not
correlated this way.
(I made the same mistake on my initial bbr testing)
=E2=80=8BBew= are of self synchronization...=C2=A0 In other situations, there have been s= ituations in which aggregate behavior occurs due to this phenomena.

So both synchronized and unsyn= chronized startup of flows under test is actually a good idea (and watching= to see if anything in the aggregate behavior over time results in such syn= chronization).
= =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 - Jim

--000000000000f7cd89056e72b361--