From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ie0-x22e.google.com (mail-ie0-x22e.google.com [IPv6:2607:f8b0:4001:c03::22e]) (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 3201021F231 for ; Mon, 7 Jul 2014 21:10:50 -0700 (PDT) Received: by mail-ie0-f174.google.com with SMTP id rd18so4537756iec.19 for ; Mon, 07 Jul 2014 21:10:48 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=IWNZGNQjADKPHIT1sdvtWJNM+tXn6zaezutuwRm9wQ0=; b=rfqffPJasqPQQfYKLPW4phZ4qJc9vZy5UM/2Fx5ksMXOtWPJ5Hyu4oCLFw7cd8tWRy 8MDiGEXrjQ3/sbtyTQyfAR5ElQ8hia/cktGS0wwtvvYu4xLFV8AQ3XIjMXrVxxe3QPTT 31afzR+ruVaM0m5dN6OGFMg1s/OyYsErzylxWtCqjQHVDk2MQploOJVqAGvSVg9S5oy3 wQgE/knyIyzoFUGkFGg9kCYc9wGHo0GMCQ8cKEPJk3+w+jKStvf/yt+DvoxSrLfp+M/4 Xk9TpNNZiyrU1UFMVwourHRg6GnEQwEeGXSAYeJik+lpVBJPUbT+s/wd7mRzJkAj03D3 fqMw== MIME-Version: 1.0 X-Received: by 10.50.4.5 with SMTP id g5mr878327igg.14.1404792648160; Mon, 07 Jul 2014 21:10:48 -0700 (PDT) Received: by 10.64.59.170 with HTTP; Mon, 7 Jul 2014 21:10:48 -0700 (PDT) Date: Mon, 7 Jul 2014 21:10:48 -0700 Message-ID: From: Aaron Wood To: bloat Content-Type: multipart/alternative; boundary=001a11c32c56c6ded804fda6c65d Subject: [Bloat] fq_codel, high bandwidth, and delays 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, 08 Jul 2014 04:10:51 -0000 --001a11c32c56c6ded804fda6c65d Content-Type: text/plain; charset=UTF-8 List, In talking with a friend over the weekend that moves data around for the national labs (on links at rates like 10Gbps), we ended up having a rather interesting discussion about just how radically different the problem spaces are vs. what he's seen in the bufferbloat community. They have few flows, long lived, and are trying to push >1Gbps per flow, across the continent (or from Europe to the US), with inherent delays on the order of 100ms. TCP under these conditions is, from his reports, incredibly fragile, where even a tiny packet error rate stops TCP for saturating the link (since it can't tell the difference between congestion and a non-congestion-related dropped packet). And suddenly the "every packet is precious" mode of thought becomes crystal clear. Clearly they are trying to solve different problems, yet they do have congestion events, when new flows are added to the network. Has anyone used fq_codel (or it's friends) in scenarios like this? fq is fairly new (2 years?) and I can't find much about it and high bandwidth links in my searches. Given that their problems aren't those that fq is trying to solve, I wouldn't expect it, but curious to see if anyone has any research on it. Thanks, Aaron --001a11c32c56c6ded804fda6c65d Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
List,

In talking with a friend over the= weekend that moves data around for the national labs (on links at rates li= ke 10Gbps), we ended up having a rather interesting discussion about just h= ow radically different the problem spaces are vs. what he's seen in the= bufferbloat community.

They have few flows, long lived, and are trying to push= >1Gbps per flow, across the continent (or from Europe to the US), with = inherent delays on the order of 100ms. =C2=A0TCP under these conditions is,= from his reports, incredibly fragile, where even a tiny packet error rate = stops TCP for saturating the link (since it can't tell the difference b= etween congestion and a non-congestion-related dropped packet).

And suddenly the "every packet is precious" m= ode of thought becomes crystal clear.

Clearly they= are trying to solve different problems, yet they do have congestion events= , when new flows are added to the network.

Has anyone used fq_codel (or it's friends) in scena= rios like this? =C2=A0fq is fairly new (2 years?) and I can't find much= about it and high bandwidth links in my searches.

Given that their problems aren't those that fq is trying to solve, I wo= uldn't expect it, but curious to see if anyone has any research on it.<= /div>

Thanks,
Aaron
--001a11c32c56c6ded804fda6c65d--