Re: [conex] Potentially interesting ConEx problem

Matt Mathis <> Tue, 05 March 2013 22:37 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id A394C11E80E9 for <>; Tue, 5 Mar 2013 14:37:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -101.978
X-Spam-Status: No, score=-101.978 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id 7B7W4GI3R3B2 for <>; Tue, 5 Mar 2013 14:37:19 -0800 (PST)
Received: from ( [IPv6:2607:f8b0:4001:c03::22b]) by (Postfix) with ESMTP id E11A111E809C for <>; Tue, 5 Mar 2013 14:37:18 -0800 (PST)
Received: by with SMTP id 10so8603561ied.30 for <>; Tue, 05 Mar 2013 14:37:17 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=EWcfMkQQ2TOxOT/O+8WUcxUxFbknHkAPhVN6uzMCfkU=; b=F4xQPYUhu0AjAPTF3Pn1Np59ddlmG5V0jjjRwg5ZOZSyVqlBKXOiQH5DZJXMMA6Y1k tRm4KMieKETahXU4rFm+vhEMpB4VZ9C1eJuKFXVIx6b81EuVyf3ron2ZRXl2bh7K0iUh SGlnfqaRhFk32xyW3tSVGqjogbWqUwiJ8OqXotyBIyMVxZmLAWmsxFvKGedEsoiT/s/M bJL5pHKuVpzx+no84Fsg/0qYRDU7p+D9iWsqBiGCZAXZdJPvQKJEutpuxRpGb4auXbVJ mNd6epRHmiyxaoZhrNlvtSsvehvXb/CPzSLmCmdUAiNLC3GxaAJ2VdGTigQICyB4zRWh CXYw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=20120113; h=mime-version:x-received:in-reply-to:references:date:message-id :subject:from:to:cc:content-type:x-gm-message-state; bh=EWcfMkQQ2TOxOT/O+8WUcxUxFbknHkAPhVN6uzMCfkU=; b=efZXxFFj/7Dgo9FczI+CerHLhuw7odkTTDvj5iO6EblY0DnUJWlL3/IhauxSF8bb0K LCG9QYxDdodXSBzRQX/FHJELNcC0XFbyO2+Jt/3sItwZ7+dNMk3IkSr1z/6u5s5IXaws LOUP1fWgJ/Qjy+fBn7VzED9zgk537dbibHtuGC7ArR5aV9sPdHCtF8TBa+Y+KsPYEUVP m4rd095NX/fq4SS6t+9LWHuBgdHDroUhp0mo27jctb6wdTrGAz8cusUl+7gxmzAZ7qUP yAZWZwfIAjJ9gxvz5CCWUaA4Wka4MyzJ2bB4X/VFz4BhHWpSdqVgyI5djVzD9JG/CoXW x6Hw==
MIME-Version: 1.0
X-Received: by with SMTP id r9mr8046671igj.96.1362523037736; Tue, 05 Mar 2013 14:37:17 -0800 (PST)
Received: by with HTTP; Tue, 5 Mar 2013 14:37:17 -0800 (PST)
In-Reply-To: <>
References: <> <> <> <>
Date: Tue, 05 Mar 2013 14:37:17 -0800
Message-ID: <>
From: Matt Mathis <>
To: "Andrew McLachlan (amclachl)" <>
Content-Type: text/plain; charset="ISO-8859-1"
X-Gm-Message-State: ALoCoQm6KnYUkTgEp2kJYbFXd5u+h7Qj1dwrgMviekaO34o0uEBHFib7QpQ2M5k78xv8kJSQ+yfrHgXQzXI98XCnr4/4Jkps5c2PIQiAxn8znTzNmflp2druWFD7BoWxu0d8Pt7Pns7JdW54jMsxIoKTWNYCBoK5VE+gYsCpP6NHL4D7xoM4xmOvJDM8vL0RzEX2BSFB2AZt
Cc: ConEx IETF list <>
Subject: Re: [conex] Potentially interesting ConEx problem
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Congestion Exposure working group discussion list <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Tue, 05 Mar 2013 22:37:20 -0000

On Tue, Mar 5, 2013 at 10:13 AM, Andrew McLachlan (amclachl)
<> wrote:
> Is this problem not mostly negated at the application layer in most video clients? They dynamically adjust to the realtime end to end bandwidth on an end user by end user basis.

The clients are not immutable, nor does the server have to tell the
clients about every available format.

> For a metro or backbone topology failure this method once again adapts per endpoint affected, over any aggregate path these clients are running over.

Right, but if I am serving from 2 mS away and the data rate is only
128kb/s, my expected TCP equilibrium loss rate might be well above 10%
(it depends on the queue size).  It makes a huge operational
difference to know if the bottleneck is one subscriber's access link
or 100k busy users trying to share a very saturated 10Gb/s link.  If
it is the latter, you really don't want to be one of the other 100k

> Moreover the expression of congestion by an ISP to an off net provider is not really an attribute you would like to advertise and used to berate the SP with.

Perhaps true, but the risks flow in both directions, and the
alternative to sharing information has the potential to have much
worse outcomes for both parties.  Besides, some inference techniques
will always work.

> QoS is your friend here if you wish to actually protect valued traffic, rather the bringing whole bunches of users to the lowest common bandwidth for an app.

Lets think about that idea: unconditionally mark all of my traffic to
be lower priority than my competitors...  I don't think that will pass
muster.   There are additional more incremental solutions in this
space, including weighted congestion control, with the ability to trim
the weights by small amounts in real time.  (Which will indirectly
cause clients to choose lower rate formats).

The question I was asking was can ConEx help the content provider know
which bottleneck prevails?   Bob?

Yes, John L, knowing which hop is key.  but recording it in the ConEx
destination option seems like a ReallyBadIdea.  For one thing all
potential tags are out of scope (I don't want to know hop counts or IP
addresses towards the user, nor if the user has a bottlenecked WiFi
between the client and the access link.)   What I really want to know
is: What are the traffic aggregates/routing prefixes sharing the same

The best way to predict the future is to create it.  - Alan Kay

Privacy matters!  We know from recent events that people are using our
services to speak in defiance of unjust governments.   We treat
privacy and security as matters of life and death, because for some
users, they are.