[tcpPrague] An L4S AQM proposal at IRTF/ACM ANRW 2020

Sándor Laki <lakis@elte.hu> Tue, 18 August 2020 07:57 UTC

Return-Path: <lakis@elte.hu>
X-Original-To: tcpprague@ietfa.amsl.com
Delivered-To: tcpprague@ietfa.amsl.com
Received: from localhost (localhost []) by ietfa.amsl.com (Postfix) with ESMTP id 092963A043E for <tcpprague@ietfa.amsl.com>; Tue, 18 Aug 2020 00:57:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.898
X-Spam-Status: No, score=-1.898 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_MSPIKE_H3=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([]) by localhost (ietfa.amsl.com []) (amavisd-new, port 10024) with ESMTP id TzGFnpwxHlPA for <tcpprague@ietfa.amsl.com>; Tue, 18 Aug 2020 00:57:00 -0700 (PDT)
Received: from mx2.mail.elte.hu (mx2.mail.elte.hu []) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5B9EB3A043A for <tcpprague@ietf.org>; Tue, 18 Aug 2020 00:56:59 -0700 (PDT)
Received: from mailbox1.caesar.elte.hu ([]) by mx2.mail.elte.hu with esmtp (Exim) id 1k7wUJ-0001ug-Ml from <lakis@elte.hu> for <tcpprague@ietf.org>; Tue, 18 Aug 2020 09:56:57 +0200
Received: (Authenticated sender: lakis) by mailbox1.caesar.elte.hu with esmtpsa (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.84_2) (envelope-from <lakis@elte.hu>) id 1k7wTz-000853-4y for tcpprague@ietf.org; Tue, 18 Aug 2020 09:56:35 +0200
To: tcpprague@ietf.org
From: Sándor Laki <lakis@elte.hu>
Message-ID: <24010457-65f3-edf0-4949-3df19962ee45@elte.hu>
Date: Tue, 18 Aug 2020 09:56:33 +0200
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:68.0) Gecko/20100101 Thunderbird/68.11.0
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-2"; format="flowed"
Content-Transfer-Encoding: 8bit
Content-Language: en-GB
X-Antivirus: Avast (VPS 200817-8, 2020.08.17), Outbound message
X-Antivirus-Status: Clean
X-ELTE-SpamScore: -4.1
X-ELTE-SpamCheck: no
X-ELTE-SpamVersion: ELTE 3.0
X-ELTE-SpamCheck-Details: score=-4.1 required=5.0 tests=ALL_TRUSTED, BAYES_00, L_AUTH autolearn=no autolearn_force=no SpamAssassin version=3.4.2 -1.0 ALL_TRUSTED Passed through trusted hosts only via SMTP -3.0 L_AUTH Caesar auth -0.1 BAYES_00 BODY: Bayes spam probability is 0 to 1% [score: 0.0000]
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpprague/4QDzuk9WDx1d-D-IdfPuiqZOdNw>
Subject: [tcpPrague] An L4S AQM proposal at IRTF/ACM ANRW 2020
X-BeenThere: tcpprague@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: "To coordinate implementation and standardisation of TCP Prague across platforms. TCP Prague will be an evolution of DCTCP designed to live alongside other TCP variants and derivatives." <tcpprague.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/tcpprague>, <mailto:tcpprague-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpprague/>
List-Post: <mailto:tcpprague@ietf.org>
List-Help: <mailto:tcpprague-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/tcpprague>, <mailto:tcpprague-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 18 Aug 2020 07:57:03 -0000

Dear All,

we presented an alternative L4S AQM proposal at ANRW 2020 few weeks ago. 
In this study, we assume that different CCs (even scalable ones) require 
different congestion signal intensities, and thus applying the same 
ECN-marking  probability leads to significant unfairness in these cases. 
E.g., BBRv2 also implements a DCTCP-like scalable mechanism, but it is 
more "aggressive" than DCTCP (using default settings). Another problem 
we realized is that BBR's model-based CC could be misled by the AQM - 
also leading to flow unfairness. As a conclusion, this heterogeneity can 
only be handled by additional mechanisms like FQ or additional packet 
marking. Our proposal marks each packet of a flow with a value 
expressing the contribution of the packet in the total throughput of the 
flow. The AQM maintains simple value thresholds, leading to different 
ECN-marking ratios of the different flows without flow-states or 
per-flow queues.

The paper and the talk is available at 


Sándor Laki, PhD
Assistant professor
Department of Information Systems
Eötvös Loránd University
Pázmány Péter stny. 1/C
H-1117, Budapest, Hungary
Room 2.506
Web: http://lakis.web.elte.hu
Phone: +36 1 372 2869 / 8477
Cell: +36 70 374 2646

Ezt az e-mailt az Avast víruskereső szoftver átvizsgálta.