From swb@internet2.edu Thu Aug 1 00:38:17 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B009721F9DA5 for ; Thu, 1 Aug 2013 00:38:17 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.735 X-Spam-Level: X-Spam-Status: No, score=-2.735 tagged_above=-999 required=5 tests=[AWL=-0.363, BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id z8X3oPxdkBgJ for ; Thu, 1 Aug 2013 00:38:11 -0700 (PDT) Received: from db9outboundpool.messaging.microsoft.com (mail-db9lp0251.outbound.messaging.microsoft.com [213.199.154.251]) by ietfa.amsl.com (Postfix) with ESMTP id B32F921F9DD6 for ; Thu, 1 Aug 2013 00:37:50 -0700 (PDT) Received: from mail88-db9-R.bigfish.com (10.174.16.254) by DB9EHSOBE036.bigfish.com (10.174.14.99) with Microsoft SMTP Server id 14.1.225.22; Thu, 1 Aug 2013 07:37:49 +0000 Received: from mail88-db9 (localhost [127.0.0.1]) by mail88-db9-R.bigfish.com (Postfix) with ESMTP id 6E2D53E0072; Thu, 1 Aug 2013 07:37:49 +0000 (UTC) X-Forefront-Antispam-Report: CIP:157.56.237.101; KIP:(null); UIP:(null); IPV:NLI; H:BY2PRD0810HT005.namprd08.prod.outlook.com; RD:none; EFVD:NLI X-SpamScore: 8 X-BigFish: S8(zzbb2dI98dIzz1f42h1d77h208ch1ee6h1de0h1fdah2073h1202h1e76h1d1ah1d2ah1fc6h1082kzzz2dh2a8h668h839h946hd25he5bhf0ah1288h12a5h12a9h12bdh137ah13b6h1441h1504h1537h153bh162dh1631h1758h1765h18e1h190ch1946h19b4h19c3h19ceh1ad9h1b0ah1d0ch1d2eh1d3fh1dfeh1dffh1e1dh1f5fh1b1cn1b1bi1155h) Received: from mail88-db9 (localhost.localdomain [127.0.0.1]) by mail88-db9 (MessageSwitch) id 1375342666294185_23950; Thu, 1 Aug 2013 07:37:46 +0000 (UTC) Received: from DB9EHSMHS007.bigfish.com (unknown [10.174.16.249]) by mail88-db9.bigfish.com (Postfix) with ESMTP id 39D45300049; Thu, 1 Aug 2013 07:37:46 +0000 (UTC) Received: from BY2PRD0810HT005.namprd08.prod.outlook.com (157.56.237.101) by DB9EHSMHS007.bigfish.com (10.174.14.17) with Microsoft SMTP Server (TLS) id 14.16.227.3; Thu, 1 Aug 2013 07:37:42 +0000 Received: from dhcp-1668.meeting.ietf.org (130.129.22.104) by pod51019.outlook.com (10.255.94.40) with Microsoft SMTP Server (TLS) id 14.16.341.1; Thu, 1 Aug 2013 07:37:40 +0000 Message-ID: <51FA1041.4010300@internet2.edu> Date: Thu, 1 Aug 2013 09:37:37 +0200 From: Scott Brim Organization: Internet2 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:17.0) Gecko/20130620 Thunderbird/17.0.7 MIME-Version: 1.0 To: "Scheffenegger, Richard" References: <012C3117EDDB3C4781FD802A8C27DD4F25C465FE@SACEXCMBX02-PRD.hq.netapp.com> In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25C465FE@SACEXCMBX02-PRD.hq.netapp.com> X-Enigmail-Version: 1.5.2 Content-Type: text/plain; charset="windows-1252" X-Originating-IP: [130.129.22.104] Content-Transfer-Encoding: quoted-printable X-OriginatorOrg: internet2.edu X-FOPE-CONNECTOR: Id%0$Dn%*$RO%0$TLS%0$FQDN%$TlsDn% Cc: "aqm@ietf.org" Subject: Re: [aqm] Draft AQM BoF minutes X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 01 Aug 2013 07:38:17 -0000 On 07/31/13 20:37, Scheffenegger, Richard allegedly wrote: > There have been comments by a few people who haven=92t given their name= s > or where their names were hard to understand. Please let me know if you > know one of them making comments so that we can update the notes. At and below "-: We heard that RED can be complicated", that's Anurag Bhargava From rs@netapp.com Thu Aug 1 01:46:43 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5558A21F87BB for ; Thu, 1 Aug 2013 01:46:43 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.729 X-Spam-Level: X-Spam-Status: No, score=-3.729 tagged_above=-999 required=5 tests=[AWL=-1.357, BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0yDBZS--QlR1 for ; Thu, 1 Aug 2013 01:46:39 -0700 (PDT) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 3FDF321F8E8E for ; Thu, 1 Aug 2013 01:46:39 -0700 (PDT) X-IronPort-AV: E=Sophos;i="4.89,793,1367996400"; d="scan'208";a="38108283" Received: from vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) by mx11-out.netapp.com with ESMTP; 01 Aug 2013 01:46:38 -0700 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.107]) by vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) with mapi id 14.03.0123.003; Thu, 1 Aug 2013 01:46:37 -0700 From: "Scheffenegger, Richard" To: Scott Brim Thread-Topic: [aqm] Draft AQM BoF minutes Thread-Index: Ac6OHPVQsch1AwBiTfqVX6bX6hLy1wAp7VyAAAxGknA= Date: Thu, 1 Aug 2013 08:46:37 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25C46BD5@SACEXCMBX02-PRD.hq.netapp.com> References: <012C3117EDDB3C4781FD802A8C27DD4F25C465FE@SACEXCMBX02-PRD.hq.netapp.com> <51FA1041.4010300@internet2.edu> In-Reply-To: <51FA1041.4010300@internet2.edu> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.118] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "aqm@ietf.org" Subject: Re: [aqm] Draft AQM BoF minutes X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 01 Aug 2013 08:46:43 -0000 Thanks a lot, updated the minutes.=20 http://www.ietf.org/proceedings/87/minutes/minutes-87-aqm This leaves 4 or 5 unidentified commenters... If you are one of them, pleas= e speak up! Richard Scheffenegger > -----Original Message----- > From: Scott Brim [mailto:swb@internet2.edu] > Sent: Donnerstag, 01. August 2013 09:38 > To: Scheffenegger, Richard > Cc: aqm@ietf.org > Subject: Re: [aqm] Draft AQM BoF minutes >=20 > On 07/31/13 20:37, Scheffenegger, Richard allegedly wrote: > > There have been comments by a few people who haven't given their names > > or where their names were hard to understand. Please let me know if > > you know one of them making comments so that we can update the notes. >=20 > At and below "-: We heard that RED can be complicated", that's Anurag > Bhargava >=20 From Martin.Stiemerling@neclab.eu Thu Aug 8 07:37:17 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7FCD511E81DC for ; Thu, 8 Aug 2013 07:37:17 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -103.481 X-Spam-Level: X-Spam-Status: No, score=-103.481 tagged_above=-999 required=5 tests=[AWL=0.118, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id h5htyhT5L2Ao for ; Thu, 8 Aug 2013 07:37:13 -0700 (PDT) Received: from mailer1.neclab.eu (mailer1.neclab.eu [195.37.70.40]) by ietfa.amsl.com (Postfix) with ESMTP id 555DB11E81D1 for ; Thu, 8 Aug 2013 07:37:11 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by mailer1.neclab.eu (Postfix) with ESMTP id 01A2A1052A0 for ; Thu, 8 Aug 2013 16:36:20 +0200 (CEST) X-Virus-Scanned: Amavisd on Debian GNU/Linux (netlab.nec.de) Received: from mailer1.neclab.eu ([127.0.0.1]) by localhost (atlas-a.office.hd [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GQnVoapD7Uwo for ; Thu, 8 Aug 2013 16:36:19 +0200 (CEST) Received: from ENCELADUS.office.hd (enceladus.office.hd [192.168.24.52]) by mailer1.neclab.eu (Postfix) with ESMTP id D55AC10529D for ; Thu, 8 Aug 2013 16:36:14 +0200 (CEST) Received: from [10.1.99.64] (10.1.99.64) by skoll.office.hd (192.168.125.11) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 8 Aug 2013 16:37:04 +0200 Message-ID: <5203AD0F.5030907@neclab.eu> Date: Thu, 8 Aug 2013 16:37:03 +0200 From: Martin Stiemerling User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130803 Thunderbird/17.0.8 MIME-Version: 1.0 To: Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [10.1.99.64] Subject: [aqm] Status after the IETF X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Aug 2013 14:37:17 -0000 Dear all, Just to let you know that the AQM BOF proposal is now in the datatracker for the IESG's internal review, i.e., it is a proposed WG by now: https://datatracker.ietf.org/doc/charter-ietf-aqm/ You can find the current proposed charter and milestones in the datatracker page. Regards, Martin -- IETF Transport Area Director martin.stiemerling@neclab.eu NEC Laboratories Europe NEC Europe Limited Registered Office: Athene, Odyssey Business Park, West End Road, London, HA4 6QE, GB Registered in England 2832014 From andrea.francini@alcatel-lucent.com Thu Aug 8 09:12:30 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CD2E421E80A8 for ; Thu, 8 Aug 2013 09:12:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -10.599 X-Spam-Level: X-Spam-Status: No, score=-10.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oFjswCwlZCxX for ; Thu, 8 Aug 2013 09:12:23 -0700 (PDT) Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id DD18D21E809E for ; Thu, 8 Aug 2013 09:12:22 -0700 (PDT) Received: from us70tusmtp1.zam.alcatel-lucent.com (h135-5-2-63.lucent.com [135.5.2.63]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id r78GCJjA004027 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL) for ; Thu, 8 Aug 2013 11:12:19 -0500 (CDT) Received: from US70UWXCHHUB02.zam.alcatel-lucent.com (us70uwxchhub02.zam.alcatel-lucent.com [135.5.2.49]) by us70tusmtp1.zam.alcatel-lucent.com (GMO) with ESMTP id r78GCIfo030002 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for ; Thu, 8 Aug 2013 12:12:18 -0400 Received: from US70TWXCHMBA12.zam.alcatel-lucent.com ([169.254.6.98]) by US70UWXCHHUB02.zam.alcatel-lucent.com ([135.5.2.49]) with mapi id 14.02.0247.003; Thu, 8 Aug 2013 12:12:18 -0400 From: "Francini, Andrea (Andrea)" To: "aqm@ietf.org" Thread-Topic: Comment about Proposed Charter Thread-Index: Ac6UUgiB/DK09jFBQRypDLLtpfDmtw== Date: Thu, 8 Aug 2013 16:12:18 +0000 Message-ID: <1BFAC0A1D7955144A2444E902CB628F80A4451@US70TWXCHMBA12.zam.alcatel-lucent.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [135.5.27.18] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39 Subject: [aqm] Comment about Proposed Charter X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Aug 2013 16:12:30 -0000 A comment about the WG proposed charter as posted at https://datatracker.ie= tf.org/doc/charter-ietf-aqm/: Where it says "minimize standing queues", I would expect also a reference t= o "maximize [or retain/protect/not compromise] link utilization". Otherwis= e any queue management scheme that keeps the standing queue small (and some= times empty) without being unfair and without causing global synchronizatio= n could be considered acceptable irrespective of the drop in link utilizati= on (-1%, -10%, -20%, ...) that it induces by occasionally emptying the queu= e (e.g., think of a RED instance configured with a maximum drop probability= that is too large for the traffic mix that it must handle).=20 Andrea From john@jlc.net Thu Aug 8 14:59:10 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 482F211E823E for ; Thu, 8 Aug 2013 14:59:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.436 X-Spam-Level: X-Spam-Status: No, score=-106.436 tagged_above=-999 required=5 tests=[AWL=-0.063, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qxbq3DvawZ1K for ; Thu, 8 Aug 2013 14:59:01 -0700 (PDT) Received: from mailhost.jlc.net (mailhost.jlc.net [199.201.159.4]) by ietfa.amsl.com (Postfix) with ESMTP id D12D921F9D38 for ; Thu, 8 Aug 2013 14:57:24 -0700 (PDT) Received: by mailhost.jlc.net (Postfix, from userid 104) id 4D31E33C22; Thu, 8 Aug 2013 17:57:15 -0400 (EDT) Date: Thu, 8 Aug 2013 17:57:15 -0400 From: John Leslie To: "Francini, Andrea (Andrea)" Message-ID: <20130808215715.GF3328@verdi> References: <1BFAC0A1D7955144A2444E902CB628F80A4451@US70TWXCHMBA12.zam.alcatel-lucent.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1BFAC0A1D7955144A2444E902CB628F80A4451@US70TWXCHMBA12.zam.alcatel-lucent.com> User-Agent: Mutt/1.4.1i Cc: "aqm@ietf.org" Subject: Re: [aqm] Comment about Proposed Charter X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Aug 2013 21:59:11 -0000 Francini, Andrea (Andrea) wrote: > > A comment about the WG proposed charter as posted at > https://datatracker.ietf.org/doc/charter-ietf-aqm/: (NB: on IESG agenda for 29 August) > Where it says "minimize standing queues", I would expect also a reference > to "maximize [or retain/protect/not compromise] link utilization". There indeed is a balancing act -- but it needs to be tilted in favor of limiting time-in-queue. > Otherwise any queue management scheme that keeps the standing queue > small (and sometimes empty) Indeed, simply dropping every packet that arrives while a previous packet isn't still being sent would "minimize standing queues" ;^( > without being unfair and without causing global synchronization could > be considered acceptable irrespective of the drop in link utilization > (-1%, -10%, -20%, ...)... I suspect that our intent is to allow ~ 1% drop in utilization; and I also suspect that anything which dropped utilization by 10% would not be deployed near the backbone. -- John Leslie From john@jlc.net Thu Aug 8 15:02:46 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6400621F9C13 for ; Thu, 8 Aug 2013 15:02:46 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.426 X-Spam-Level: X-Spam-Status: No, score=-106.426 tagged_above=-999 required=5 tests=[AWL=-0.054, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, SARE_SUB_OBFU_Q1=0.227, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NGpkGahzBlmV for ; Thu, 8 Aug 2013 15:02:40 -0700 (PDT) Received: from mailhost.jlc.net (mailhost.jlc.net [199.201.159.4]) by ietfa.amsl.com (Postfix) with ESMTP id 5C69521F9C0E for ; Thu, 8 Aug 2013 15:02:33 -0700 (PDT) Received: by mailhost.jlc.net (Postfix, from userid 104) id 03B7633C21; Thu, 8 Aug 2013 18:02:33 -0400 (EDT) Date: Thu, 8 Aug 2013 18:02:32 -0400 From: John Leslie To: Martin Stiemerling Message-ID: <20130808220232.GG3328@verdi> References: <5203AD0F.5030907@neclab.eu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5203AD0F.5030907@neclab.eu> User-Agent: Mutt/1.4.1i Cc: aqm@ietf.org Subject: Re: [aqm] Status after the IETF X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Aug 2013 22:02:46 -0000 Martin Stiemerling wrote: > > Just to let you know that the AQM BOF proposal is now in the datatracker > for the IESG's internal review, i.e., it is a proposed WG by now: > https://datatracker.ietf.org/doc/charter-ietf-aqm/ I note desideratum (2): " " help flow sources control their sending rates without unnecessary losses, " e.g. through ECN Is this just "motherhood and apple pie" or is there an intent to allow alternatives/modifications to ECN which _wouldn't_ be rejected by nearly all operators? -- John Leslie From rs@netapp.com Fri Aug 9 01:59:00 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EAEE821F9F0A for ; Fri, 9 Aug 2013 01:59:00 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.887 X-Spam-Level: X-Spam-Status: No, score=-3.887 tagged_above=-999 required=5 tests=[AWL=-1.515, BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iF2YSFoTgBqY for ; Fri, 9 Aug 2013 01:58:56 -0700 (PDT) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id E2C0421F9E63 for ; Fri, 9 Aug 2013 01:58:56 -0700 (PDT) X-IronPort-AV: E=Sophos;i="4.89,844,1367996400"; d="scan'208";a="40344598" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx11-out.netapp.com with ESMTP; 09 Aug 2013 01:58:54 -0700 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.48]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0123.003; Fri, 9 Aug 2013 01:58:53 -0700 From: "Scheffenegger, Richard" To: John Leslie , Martin Stiemerling Thread-Topic: [aqm] Status after the IETF Thread-Index: AQHOlETyQn1xxH5YR0eYo6aA+VirqJmMUsUAgABBBGA= Date: Fri, 9 Aug 2013 08:58:53 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25C8D2ED@SACEXCMBX02-PRD.hq.netapp.com> References: <5203AD0F.5030907@neclab.eu> <20130808220232.GG3328@verdi> In-Reply-To: <20130808220232.GG3328@verdi> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.115] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "aqm@ietf.org" Subject: Re: [aqm] Status after the IETF X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Aug 2013 08:59:01 -0000 Hi John, this was shortly discussed. If you are thinking about changing the semantic= s of ECN, a different signaling scheme for congestion, etc, that would be c= onsidered out-of-scope for the AQM WG (TSVWG would be the place to take thi= s). However, an AQM utilizing such a modification / extension to the congestion= signaling would be in-charter, but any draft with such a dependency would = probably take a long time to pass... So, the charter doesn't explicitly exc= lude this, but at this point, it's hard to envision a combined ECN modifica= tion / AQM scheme that could find fast support. After all, there are many kinds of different signaling / AQM schemes one ca= n come up with, if designing on a green field... Richard Scheffenegger > -----Original Message----- > From: aqm-bounces@ietf.org [mailto:aqm-bounces@ietf.org] On Behalf Of Joh= n > Leslie > Sent: Freitag, 09. August 2013 00:03 > To: Martin Stiemerling > Cc: aqm@ietf.org > Subject: Re: [aqm] Status after the IETF >=20 > Martin Stiemerling wrote: > > > > Just to let you know that the AQM BOF proposal is now in the > > datatracker for the IESG's internal review, i.e., it is a proposed WG b= y > now: > > https://datatracker.ietf.org/doc/charter-ietf-aqm/ >=20 > I note desideratum (2): > " > " help flow sources control their sending rates without unnecessary > losses, " e.g. through ECN >=20 > Is this just "motherhood and apple pie" or is there an intent to allow > alternatives/modifications to ECN which _wouldn't_ be rejected by nearly > all operators? >=20 > -- > John Leslie > _______________________________________________ > aqm mailing list > aqm@ietf.org > https://www.ietf.org/mailman/listinfo/aqm From swmike@swm.pp.se Fri Aug 9 09:01:37 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EC82321F9F0A for ; Fri, 9 Aug 2013 09:01:37 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.372 X-Spam-Level: X-Spam-Status: No, score=-2.372 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6L5dMNfXQoL9 for ; Fri, 9 Aug 2013 09:01:37 -0700 (PDT) Received: from uplift.swm.pp.se (ipv6.swm.pp.se [IPv6:2a00:801::f]) by ietfa.amsl.com (Postfix) with ESMTP id 7A9F511E8126 for ; Fri, 9 Aug 2013 08:55:11 -0700 (PDT) Received: by uplift.swm.pp.se (Postfix, from userid 501) id 89650A1; Fri, 9 Aug 2013 17:55:09 +0200 (CEST) Received: from localhost (localhost [127.0.0.1]) by uplift.swm.pp.se (Postfix) with ESMTP id 850049A; Fri, 9 Aug 2013 17:55:09 +0200 (CEST) Date: Fri, 9 Aug 2013 17:55:09 +0200 (CEST) From: Mikael Abrahamsson To: "Scharf, Michael (Michael)" In-Reply-To: <655C07320163294895BBADA28372AF5D07830B@FR712WXCHMBA15.zeu.alcatel-lucent.com> Message-ID: References: <51F6633B.8000509@mti-systems.com> <201307291650.r6TGojBh009833@bagheera.jungle.bt.co.uk> <655C07320163294895BBADA28372AF5D07830B@FR712WXCHMBA15.zeu.alcatel-lucent.com> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) Organization: People's Front Against WWW MIME-Version: 1.0 Content-Type: TEXT/PLAIN; format=flowed; charset=US-ASCII Cc: "aqm@ietf.org" Subject: Re: [aqm] AQM BoF - RTGAREA slides X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Aug 2013 16:01:38 -0000 On Tue, 30 Jul 2013, Scharf, Michael (Michael) wrote: > I am just curious: Is there any data that shows *any* benefit of RED (or > whatever other AQM flavor) on a 10G or 100G link? As opposed to FIFO? I'd say that if you made a platform with 600ms buffer and tried to cram 130 gigabit/s of data through it, I'm sure RED would be of benefit. Not a data point, but I have some experience with (accidentally) saturating links with a lot of traffic. For instance a 1G link (10 years ago when gig was fast) that when it was flatlining didn't have more than approximately 5-10ms worth of buffers, then would tail drop. This was still usable when congesting, when (later) there was approximately 1.3 gigabit/s of traffic after upgrade to 10GE, so approximately 30% backpressure. I've seen 1G links on "real" core routers with 600ms buffers and back in the days with mostly Win98, Win2k and XP clients you would see 10-20 ms of buffers because the clients didn't have more than 16-64k TCP windows, so it would hold stable at 10-20ms buffering and zero loss. Current core routers seem to have converged onto around 50ms of buffers and I think this makes a lot of sense. If I had a bit more time and equipment I'd really like to test a theory of mine, and that is to try to run thousands of TCP sessions through a 100GE link (10GE might be enough) with RED set at 0% loss 0-5ms buffer depth drop and then from 5-50 ms buffer depth ramp up from 0 to 100%. I woud also like to experiment with that 100% at 50ms set to decrements of 10. My theory is that 10% drop probability at 50ms depth might be enough to make TCP never reach this point in a sustained fashion. -- Mikael Abrahamsson email: swmike@swm.pp.se From john@jlc.net Fri Aug 9 09:54:05 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8499D21F9D2A for ; Fri, 9 Aug 2013 09:54:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.533 X-Spam-Level: X-Spam-Status: No, score=-106.533 tagged_above=-999 required=5 tests=[AWL=0.066, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aM6vo8xyQFgw for ; Fri, 9 Aug 2013 09:53:56 -0700 (PDT) Received: from mailhost.jlc.net (mailhost.jlc.net [199.201.159.4]) by ietfa.amsl.com (Postfix) with ESMTP id AD39A21F9D53 for ; Fri, 9 Aug 2013 09:47:30 -0700 (PDT) Received: by mailhost.jlc.net (Postfix, from userid 104) id 4B2CD33C27; Fri, 9 Aug 2013 12:47:29 -0400 (EDT) Date: Fri, 9 Aug 2013 12:47:29 -0400 From: John Leslie To: "Scheffenegger, Richard" Message-ID: <20130809164729.GI3328@verdi> References: <5203AD0F.5030907@neclab.eu> <20130808220232.GG3328@verdi> <012C3117EDDB3C4781FD802A8C27DD4F25C8D2ED@SACEXCMBX02-PRD.hq.netapp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25C8D2ED@SACEXCMBX02-PRD.hq.netapp.com> User-Agent: Mutt/1.4.1i Cc: aqm@ietf.org, Martin Stiemerling Subject: [aqm] ECN language in the proposed charter X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Aug 2013 16:54:05 -0000 Scheffenegger, Richard wrote: > To: John Leslie > > >> I note desideratum (2): >> " >> " help flow sources control their sending rates without unnecessary >> " losses, " e.g. through ECN >> >> Is this just "motherhood and apple pie" or is there an intent to allow >> alternatives/modifications to ECN which _wouldn't_ be rejected by nearly >> all operators? > > this was shortly discussed. If you are thinking about changing the > semantics of ECN, a different signaling scheme for congestion, etc, > that would be considered out-of-scope for the AQM WG (TSVWG would be > the place to take this). No doubt... but what is this language doing in an AQM charter, then? > However, an AQM utilizing such a modification / extension to the > congestion signaling would be in-charter, but any draft with such > a dependency would probably take a long time to pass... Sounds like you're recommending against this, absent a recharter... > So, the charter doesn't explicitly exclude this, but at this point, > it's hard to envision a combined ECN modification / AQM scheme that > could find fast support. Define "fast"... > After all, there are many kinds of different signaling / AQM schemes > one can come up with, if designing on a green field... Actually, there's no need for a "green field": anything a forwarding node does is so far de-coupled from what a transport sender does as to really belong in separate standards. The fatal problem with the ECN standard is that it tries to combine what a node does with what a transport does. (The problem created by ] ] ... MUST be essentially the same as the congestion control response ] to a *single* dropped packet is a lesser problem, more easily corrected.) ECN as described in RFC 3168 has been un-deployable since day one: because the transport can't know what forwarding nodes may do, and forwarding nodes can't know what the transport may do. Folks have dutifully implemented it, both in transports and in forwarding stacks, but have found themselves forced to turn it off by default. To be deployable, we need something where forwarding nodes have no reason to worry what transports may do with it. Likewise, transports must be allowed to guess at what forwarding nodes mean by it. The essential point of ECN is that it's _explicit_ -- thus it requires only one RTT to get back to the sender. "So what?" too many folks ask when I say that. :^( Anything that comes back _faster_ than you can infer a drop gives better information -- thus, it _should_ allow a faster but less drastic response. Generic TCP starts in "slow-start" mode, with an obviously-bad guess for an Initial Window, increasing that each time an ACK is received. Thus, it turns out to be exponential growth in _time_. This is _bad_! It nearly guarantees overshoot by a factor of two when a loss has to be inferred. OTOH, we now have real-time flows -- what AQM is being chartered to consider -- that have no desire to grow exponentially. They need to co-exist with large-transfer TCP sessions which are happy to grow exponentially. The signal a transport _wants_ from forwarding nodes is that we've reached the time at which exponential growth of sending rate deserves to stop. (Obviously, we can't receive it at the same moment; but a transport has enough information to make a good guess at what that moment corresponds to in its sending history.) The reaction to such a signal should be to stop the exponential growth and cut the sending rate back to where it was when the signal was generated. For bulk-transfer TCP, this would be a bit higher than the current algorithm; but for a real-time stream it might be almost no reduction. The necessary _and_ sufficient change to ECN for forwarding nodes is that the ECN signal be given _before_ packets need to be dropped -- it really doesn't have to be _much_ before. Granted, 3168 explicitly forbids that now; but so long as the actual behavior is documented, any TCP stack can compensate -- and in many cases, no significant compensation is needed. Once the forwarding stacks signal ECN _before_ drop, all worries about senders using ECN to game the system (at forwarding nodes) disappear. ==== Why am I posting this here, instead of in TSVVG? Simply, because it needs to be considered by people who _care_! But beyond that, I believe the work of modifying ECN needs to be _done_ by people who care -- and this looks to be the place to find them. I am willing to lobby Martin and Spencer; and I'm willing to run interference when the charter goes to External Review. The first question needs to be "Is this worth doing?" rather than "How do we make folks deploy it?" -- John Leslie From naeem.khademi@gmail.com Wed Aug 14 07:05:17 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0AACF21E8097 for ; Wed, 14 Aug 2013 07:05:17 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.372 X-Spam-Level: X-Spam-Status: No, score=-2.372 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Kaz4RrjVkN9l for ; Wed, 14 Aug 2013 07:05:16 -0700 (PDT) Received: from mail-ve0-x22d.google.com (mail-ve0-x22d.google.com [IPv6:2607:f8b0:400c:c01::22d]) by ietfa.amsl.com (Postfix) with ESMTP id D3D5711E814C for ; Wed, 14 Aug 2013 07:05:15 -0700 (PDT) Received: by mail-ve0-f173.google.com with SMTP id cy12so7632994veb.4 for ; Wed, 14 Aug 2013 07:05:15 -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=/H8D7nSoJJmB4dKReHIupZzo2fTF3FI4VXK+doUIOBo=; b=pjF92Hxluiclzu40OK2ZzO5tHk2tmNWk5f638DODNcHlf40oDcZJga76MnucVJtQSL HHbUICB/r8EvgF+veq3yjLdsvsJsPBWSNKnKFQMkqAw/pWJYp7T2+f8baepdlx/IYbOu kTyHKkmvZMbli2VVNWT6Zu0AaJ+Mi0T0acZYYduMUM19FGLhQ5QHPtPNnLcL4HsDolss DDxLUYAbOy/zdSdr88WRhs9QpzVNw8avy06JBs76fIhQN9sFBLLdSGcS/xNP3/yd6ZJG 73N1UF0qEQtCBdy315Ouum1H+Fmm8ftwm8WFc79yW+jn0RMnckO7cZZ8FJJ/WjbZKo+p NkRg== MIME-Version: 1.0 X-Received: by 10.58.211.227 with SMTP id nf3mr9907996vec.20.1376489104760; Wed, 14 Aug 2013 07:05:04 -0700 (PDT) Received: by 10.221.51.198 with HTTP; Wed, 14 Aug 2013 07:05:04 -0700 (PDT) Date: Wed, 14 Aug 2013 16:05:04 +0200 Message-ID: From: Naeem Khademi To: aqm@ietf.org, iccrg@irtf.org, bloat@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=047d7bd6b65a20659204e3e8d85b Subject: [aqm] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 14 Aug 2013 14:05:17 -0000 --047d7bd6b65a20659204e3e8d85b Content-Type: text/plain; charset=ISO-8859-1 Hi The already-existing PIE Linux code maintained at ftp://ftpeng.cisco.com/pie/ does not support ECN. Studying the interaction of ECN and AQM algos is probably of interest of some of us. I have added ECN support to PIE Linux code and have given it some basic sanity tests. The modified code is available at http://heim.ifi.uio.no/naeemk/research/PIE/ I would appreciate any feedback. Regards, Naeem --047d7bd6b65a20659204e3e8d85b Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Hi=A0

The already-existing PIE Li= nux code maintained at=A0ftp://ftpe= ng.cisco.com/pie/=A0does not support ECN. Studying the interaction of E= CN and AQM algos is probably of interest of some of us. I have added ECN su= pport to PIE Linux code and have given it some basic sanity tests.=A0

The modified code is available at=A0http://heim.ifi.uio.no/nae= emk/research/PIE/

I would apprecia= te any feedback.=A0

Regards,
Naeem


--047d7bd6b65a20659204e3e8d85b-- From subramanian.vijay@gmail.com Wed Aug 14 11:27:50 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A6BAA21E80BB for ; Wed, 14 Aug 2013 11:27:50 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.372 X-Spam-Level: X-Spam-Status: No, score=-2.372 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4sA-ZmaJ4pmY for ; Wed, 14 Aug 2013 11:27:49 -0700 (PDT) Received: from mail-qe0-x235.google.com (mail-qe0-x235.google.com [IPv6:2607:f8b0:400d:c02::235]) by ietfa.amsl.com (Postfix) with ESMTP id 5D10B11E8172 for ; Wed, 14 Aug 2013 11:27:49 -0700 (PDT) Received: by mail-qe0-f53.google.com with SMTP id 1so1066820qee.40 for ; Wed, 14 Aug 2013 11:27:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=vVPPiSfIcJCBXJxVHkd4kMu285ypxihvZsiTEhRAqHw=; b=cdsdEaFsle5GUwrIh3N/IMJFIViuxDIive5/5TDJb+zainOuprq1jLrQx+JTZ3aBgB 9aLtse4LRx7Ts3JWorP2OErEpnd8gakSqCQVrSQivGw0M7CeDESiLxD83yK3YMNWMx6x fRCaW4nBnDDcqiTAt1mooFnnSNI9yuBHEOMPVBuptf3Bg3r6/7dQjmsdx/lVV6EXXKyW QuxfkJjfTiZ/19Bhk0MMhhEYYhxSdabZ/mbohDCpATUo0qKkkT5vfAWGGNyLHq4fK8Nr BD6puAt/Oj50X+UJ0BBa4wBkzNwp2T/IqZ+gpyiR3v+pXq6qXhSePdVQiJm9Zdt/A5/8 gMsg== MIME-Version: 1.0 X-Received: by 10.224.113.209 with SMTP id b17mr6640322qaq.85.1376504860924; Wed, 14 Aug 2013 11:27:40 -0700 (PDT) Received: by 10.49.105.232 with HTTP; Wed, 14 Aug 2013 11:27:40 -0700 (PDT) In-Reply-To: References: Date: Wed, 14 Aug 2013 11:27:40 -0700 Message-ID: From: Vijay Subramanian To: Naeem Khademi Content-Type: multipart/alternative; boundary=047d7bdcab9044854504e3ec834b Cc: iccrg@irtf.org, "aqm@ietf.org" , bloat@lists.bufferbloat.net Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 14 Aug 2013 18:27:50 -0000 --047d7bdcab9044854504e3ec834b Content-Type: text/plain; charset=ISO-8859-1 Naeem, Thanks for testing and the contribution! We have ECN support in our current code-under-development thanks to earlier help from Dave Taht but I will merge your changes as appropriate and submit it back for review. I will take a look at your code soon. Do you also have any test results which you can share? Thanks again, Vijay On 14 August 2013 07:05, Naeem Khademi wrote: > Hi > > The already-existing PIE Linux code maintained at > ftp://ftpeng.cisco.com/pie/ does not support ECN. Studying the > interaction of ECN and AQM algos is probably of interest of some of us. I > have added ECN support to PIE Linux code and have given it some basic > sanity tests. > > The modified code is available at > http://heim.ifi.uio.no/naeemk/research/PIE/ > > I would appreciate any feedback. > > Regards, > Naeem > > > > _______________________________________________ > iccrg mailing list > iccrg@irtf.org > https://www.irtf.org/mailman/listinfo/iccrg > > --047d7bdcab9044854504e3ec834b Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Naeem,

Thanks for testing and the contr= ibution! We have ECN support in our current code-under-development thanks t= o earlier help from Dave Taht but I will merge your changes as appropriate = and
submit it back for review.=A0

I will take a l= ook at your code soon. Do you also have any test results which you can shar= e?=A0

Thanks again,
Vijay


On 14 August 2013 07:05, Naeem Khademi <= span dir=3D"ltr"><naeem.khademi@gmail.com> wrote:
Hi=A0

The already-existing PIE Linux co= de maintained at=A0ftp://ftpeng.cisco.com/pie/=A0does not support ECN. Studying the inte= raction of ECN and AQM algos is probably of interest of some of us. I have = added ECN support to PIE Linux code and have given it some basic sanity tes= ts.=A0

The modified code is available at=A0http://heim.ifi.uio.= no/naeemk/research/PIE/

I would appreciate any= feedback.=A0

Regards,
Naeem


<= /div>

_______________________________________________
iccrg mailing list
iccrg@irtf.org
h= ttps://www.irtf.org/mailman/listinfo/iccrg


--047d7bdcab9044854504e3ec834b-- From naeem.khademi@gmail.com Wed Aug 14 14:17:27 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ED76C11E815C for ; Wed, 14 Aug 2013 14:17:26 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.373 X-Spam-Level: X-Spam-Status: No, score=-2.373 tagged_above=-999 required=5 tests=[AWL=0.000, BAYES_00=-2.599, NO_RELAYS=-0.001, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id sWaL9EJfmttG for ; Wed, 14 Aug 2013 14:17:26 -0700 (PDT) Received: from mail-vc0-x230.google.com (mail-vc0-x230.google.com [IPv6:2607:f8b0:400c:c03::230]) by ietfa.amsl.com (Postfix) with ESMTP id 67F0F11E80A2 for ; Wed, 14 Aug 2013 14:17:26 -0700 (PDT) Received: by mail-vc0-f176.google.com with SMTP id ha11so5645217vcb.35 for ; Wed, 14 Aug 2013 14:17:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=HCgyp5rsWs5VqxCAEs+3OD1IqHknD2GpWOnnjwWrXk8=; b=0JE7c8MaHU7uf5nZ2o9C0rSOLV4Th0UmFWfnumIB8rDnBL72y1FklwkzLtT30lNcCD 7IkC6zne8ucU9OB7bEquuveciNr8IYlKHdep30VZbrz9vTrYGkojsG0LkewmMYPfhqDK AuE0c7ZUFfmc0mn1RuyyYiya1Cy6QzHACaHmRv3/BSVpPevieaWk5C2414hTN+KLamf5 9LKklueKfr5i19j0in0q2pqimkm8MEDEXp7vP+aBlVd0R3nGollBDgxYPuIO8AphksUW NlUdq8xThFdkcDSaSaKPf3DRpnm9hS6gyYVsanx+D1h6SGv7Ba53wI8GAQ742nnHmHy3 15VA== MIME-Version: 1.0 X-Received: by 10.58.29.111 with SMTP id j15mr51063veh.76.1376515041067; Wed, 14 Aug 2013 14:17:21 -0700 (PDT) Received: by 10.221.51.198 with HTTP; Wed, 14 Aug 2013 14:17:21 -0700 (PDT) In-Reply-To: References: Date: Wed, 14 Aug 2013 23:17:21 +0200 Message-ID: From: Naeem Khademi To: Vijay Subramanian Content-Type: text/plain; charset=ISO-8859-1 Cc: iccrg@irtf.org, "aqm@ietf.org" , bloat@lists.bufferbloat.net Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 14 Aug 2013 21:17:27 -0000 Hi Vijay This code implements basic ECN capability for PIE meaning that it would CE-mark packets that would be early-dropped otherwise. I have sanity-checked the code (e.g. successful ECN marking and receiver/sender reaction). We're now in process of conducting larger set of experiments with PIE w/wo ECN that will eventually be published as we move forward. Cheers, Naeem On Wed, Aug 14, 2013 at 8:27 PM, Vijay Subramanian wrote: > Naeem, > > Thanks for testing and the contribution! We have ECN support in our current > code-under-development thanks to earlier help from Dave Taht but I will > merge your changes as appropriate and > submit it back for review. > > I will take a look at your code soon. Do you also have any test results > which you can share? > > Thanks again, > Vijay > > > On 14 August 2013 07:05, Naeem Khademi wrote: >> >> Hi >> >> The already-existing PIE Linux code maintained at >> ftp://ftpeng.cisco.com/pie/ does not support ECN. Studying the interaction >> of ECN and AQM algos is probably of interest of some of us. I have added ECN >> support to PIE Linux code and have given it some basic sanity tests. >> >> The modified code is available at >> http://heim.ifi.uio.no/naeemk/research/PIE/ >> >> I would appreciate any feedback. >> >> Regards, >> Naeem >> >> >> >> _______________________________________________ >> iccrg mailing list >> iccrg@irtf.org >> https://www.irtf.org/mailman/listinfo/iccrg >> > From dave.taht@gmail.com Wed Aug 14 17:42:04 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8438021E80FE for ; Wed, 14 Aug 2013 17:42:04 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.653 X-Spam-Level: X-Spam-Status: No, score=-1.653 tagged_above=-999 required=5 tests=[AWL=0.719, BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZNhDn0ItIx5p for ; Wed, 14 Aug 2013 17:42:03 -0700 (PDT) Received: from mail-we0-x232.google.com (mail-we0-x232.google.com [IPv6:2a00:1450:400c:c03::232]) by ietfa.amsl.com (Postfix) with ESMTP id 110CE21F9E76 for ; Wed, 14 Aug 2013 17:42:00 -0700 (PDT) Received: by mail-we0-f178.google.com with SMTP id u57so105697wes.23 for ; Wed, 14 Aug 2013 17:42:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=wCrl1lW5J26T1W2yGhONWV61B3g3+ZbENyDmfW4usrU=; b=PT9WG66AmteMIDf2I9shM3qcZCo6gKX7DYGmg1WUZc9vyeyswq/xxuox4+5cIUEj1X Z1+zDR8iYo988YMbfOTWGJLt/3Ji1S9BMhrM6LlELM2+RsxFnfIV09dELTnLNHeGhinH 5JXOMl7ZeRQO82FX3MiESuWpqn0Rf618ENCFk8t4vvzl4lsAT/toT/n4/w7rNgYPhzHT ECXX1v4NoeYu2YApjIOtQHiByQXeg1fgYj4rQkOTxjzJJPQsXnJASL1SAJ5O1uab2mPo PUAM/wt4fFzbAKH1Wffza77MMKfKF+Jkg0Qt7+1fG7vPBCi0gCm7UtScaMwI5jpkxUcP C1Vg== MIME-Version: 1.0 X-Received: by 10.180.74.134 with SMTP id t6mr142776wiv.56.1376527319966; Wed, 14 Aug 2013 17:41:59 -0700 (PDT) Received: by 10.217.48.129 with HTTP; Wed, 14 Aug 2013 17:41:59 -0700 (PDT) In-Reply-To: References: Date: Wed, 14 Aug 2013 17:41:59 -0700 Message-ID: From: Dave Taht To: Naeem Khademi Content-Type: multipart/alternative; boundary=f46d043c820cee2b9a04e3f1bd43 Cc: "iccrg@irtf.org" , "aqm@ietf.org" , Vijay Subramanian , bloat Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 15 Aug 2013 00:42:04 -0000 --f46d043c820cee2b9a04e3f1bd43 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable The pie patches that I was using until recently are here: http://snapon.lab.bufferbloat.net/~cero2/deb/patches/ Stephen hemminger and I cleaned up the original pie drop according to KernelStyle in multiple ways. There is a rounding error problem in setting the target value (we'd recomended they switch to ktime rather than jiffies) that I either added or was in there in the first place.... and we fixed a couple generic bugs... In addition to adding ecn support, we also added congestion notification to (a la the linux red and fq_codel algorithms) to provide lossless feedback into the host tcp stack. (this is, incidentally, 1 line of code) This optimization made it possible to do comparisons on host <-> host tests between these qdiscs, with no router in-between. Syntactically the iproute2 patches could use identical syntax for time based values as the codel iproute2 patches. Haven't got around to that. (not my job, really, to improve cisco's code) Now, I only now - today! - just noticed that the new pie code at cisco's site seems *substantially* different from the older pie code the above was based on, in that it no longer has an optimization for not dropping smaller packets. (bytemode) This is going to invalidate most previous benchmarks of pie that I know of, including all of mine, and most, if not all, of cablelabs'. Sigh. I will try to refresh my patches and update the lab over the next week or so. On Wed, Aug 14, 2013 at 2:17 PM, Naeem Khademi wro= te: > Hi Vijay > > This code implements basic ECN capability for PIE meaning that it > would CE-mark packets that would be early-dropped otherwise. I have > sanity-checked the code (e.g. successful ECN marking and > receiver/sender reaction). We're now in process of conducting larger > set of experiments with PIE w/wo ECN that will eventually be published > as we move forward. > > Cheers, > Naeem > > On Wed, Aug 14, 2013 at 8:27 PM, Vijay Subramanian > wrote: > > Naeem, > > > > Thanks for testing and the contribution! We have ECN support in our > current > > code-under-development thanks to earlier help from Dave Taht but I will > > merge your changes as appropriate and > > submit it back for review. > > > > I will take a look at your code soon. Do you also have any test results > > which you can share? > > > > Thanks again, > > Vijay > > > > > > On 14 August 2013 07:05, Naeem Khademi wrote: > >> > >> Hi > >> > >> The already-existing PIE Linux code maintained at > >> ftp://ftpeng.cisco.com/pie/ does not support ECN. Studying the > interaction > >> of ECN and AQM algos is probably of interest of some of us. I have > added ECN > >> support to PIE Linux code and have given it some basic sanity tests. > >> > >> The modified code is available at > >> http://heim.ifi.uio.no/naeemk/research/PIE/ > >> > >> I would appreciate any feedback. > >> > >> Regards, > >> Naeem > >> > >> > >> > >> _______________________________________________ > >> iccrg mailing list > >> iccrg@irtf.org > >> https://www.irtf.org/mailman/listinfo/iccrg > >> > > > _______________________________________________ > iccrg mailing list > iccrg@irtf.org > https://www.irtf.org/mailman/listinfo/iccrg > --=20 Dave T=E4ht Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html --f46d043c820cee2b9a04e3f1bd43 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
The pie patches that I was using until recently are here:<= br><= div> Stephen hemminger and I cleaned up the original pie drop according to Kerne= lStyle in multiple ways. There is a rounding error problem in setting the t= arget value
(we'd recomended they switch to ktime rather= than jiffies) that I either added or was in there in the first place.... a= nd we fixed a couple generic bugs...

In addition to adding ecn support, we also added congestion = notification to (a la the linux red and fq_codel algorithms) to provide los= sless feedback into the host tcp stack. (this is, incidentally, 1 line of c= ode) This optimization made it possible to do comparisons on host <->= host tests between these qdiscs, with no router in-between.

Syntactically the iproute2 patches could use identical synta= x for time based values as the codel iproute2 patches. Haven't got arou= nd to that. (not my job, really, to improve cisco's code)

Now, I only now - today! - just noticed that the new pie code at cisco= 's site seems *substantially* different from the older pie code the abo= ve was based on, in that it no longer has an optimization for not dropping = smaller packets. (bytemode) This is going to invalidate most previous bench= marks of pie that I know of, including all of mine, and most, if not all, o= f cablelabs'. Sigh.

I will try to refresh my patches and update the lab over the= next week or so.

=

On Wed, Aug 14, 2013 at 2:17 PM, Naeem K= hademi <naeem.khademi@gmail.com> wrote:
Hi Vijay

This code implements basic ECN capability for PIE meaning that it
would CE-mark packets that would be early-dropped otherwise. I have
sanity-checked the code (e.g. successful ECN marking and
receiver/sender reaction). We're now in process of conducting larger set of experiments with PIE w/wo ECN that will eventually be published
as we move forward.

Cheers,
Naeem

On Wed, Aug 14, 2013 at 8:27 PM, Vijay Subramanian
<subramanian.vijay@gmail.= com> wrote:
> Naeem,
>
> Thanks for testing and the contribution! We have ECN support in our cu= rrent
> code-under-development thanks to earlier help from Dave Taht but I wil= l
> merge your changes as appropriate and
> submit it back for review.
>
> I will take a look at your code soon. Do you also have any test result= s
> which you can share?
>
> Thanks again,
> Vijay
>
>
> On 14 August 2013 07:05, Naeem Khademi <naeem.khademi@gmail.com> wrote:
>>
>> Hi
>>
>> The already-existing PIE Linux code maintained at
>> ftp://ft= peng.cisco.com/pie/ does not support ECN. Studying the interaction
>> of ECN and AQM algos is probably of interest of some of us. I have= added ECN
>> support to PIE Linux code and have given it some basic sanity test= s.
>>
>> The modified code is available at
>> http://heim.ifi.uio.no/naeemk/research/PIE/
>>
>> I would appreciate any feedback.
>>
>> Regards,
>> Naeem
>>
>>
>>
>> _______________________________________________
>> iccrg mailing list
>> iccrg@irtf.org
>> https://www.irtf.org/mailman/listinfo/iccrg
>>
>
_______________________________________________
iccrg mailing list
iccrg@irtf.org
h= ttps://www.irtf.org/mailman/listinfo/iccrg



--
Dave T=E4ht=

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/= subscribe.html=20
--f46d043c820cee2b9a04e3f1bd43-- From lars@netapp.com Wed Aug 14 18:25:20 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 87B4B11E80F4 for ; Wed, 14 Aug 2013 18:25:20 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.486 X-Spam-Level: X-Spam-Status: No, score=-6.486 tagged_above=-999 required=5 tests=[AWL=-4.114, BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1U9CcX1NAcsz for ; Wed, 14 Aug 2013 18:25:15 -0700 (PDT) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 931B711E80D2 for ; Wed, 14 Aug 2013 18:25:15 -0700 (PDT) X-IronPort-AV: E=Sophos;i="4.89,881,1367996400"; d="asc'?scan'208";a="41567591" Received: from vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) by mx11-out.netapp.com with ESMTP; 14 Aug 2013 18:25:15 -0700 Received: from SACEXCMBX01-PRD.hq.netapp.com ([169.254.2.240]) by vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) with mapi id 14.03.0123.003; Wed, 14 Aug 2013 18:25:15 -0700 From: "Eggert, Lars" To: Dave Taht Thread-Topic: [aqm] [iccrg] ECN support for PIE AQM in Linux Thread-Index: AQHOmVBFFA6HspGimkyLiVr34NauMpmV700A Date: Thu, 15 Aug 2013 01:25:14 +0000 Message-ID: <2A9ED096-C647-4D7A-A32B-422F1DFC3ACA@netapp.com> References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.117] Content-Type: multipart/signed; boundary="Apple-Mail=_ECB9A846-6B3D-4A1A-84B7-045EC3B9A868"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Cc: bloat , "iccrg@irtf.org" , Naeem Khademi , Vijay Subramanian , "aqm@ietf.org" Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 15 Aug 2013 01:25:20 -0000 --Apple-Mail=_ECB9A846-6B3D-4A1A-84B7-045EC3B9A868 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=iso-8859-1 On Aug 15, 2013, at 8:41, Dave Taht wrote: > Now, I only now - today! - just noticed that the new pie code at = cisco's > site seems *substantially* different from the older pie code the above = was > based on I guess the lesson is that these code drops need version numbers, and we = need to REPORT the versions whenever we present results. Lars=20 --Apple-Mail=_ECB9A846-6B3D-4A1A-84B7-045EC3B9A868 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iQCVAwUBUgwt+tZcnpRveo1xAQKN6AQAqJ81/pSeP//Wc7veJ+vCj0lR76Z2b1WG 0wLdPnp3WiwmB75ZLj6NRxsqVYvEd1X+WmLbmOBR+tT824Cpys5ZZvNjYxvO5nhq PH/gYCKJwGmgRpPKj1WX0he7SKNxbNDAg2L/YX0/ShXyjBC0FN+1ray6QBsinCQK rRDnI0PftBs= =9YMu -----END PGP SIGNATURE----- --Apple-Mail=_ECB9A846-6B3D-4A1A-84B7-045EC3B9A868-- From ropan@cisco.com Wed Aug 14 18:42:19 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8264611E81AA for ; Wed, 14 Aug 2013 18:42:19 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -10.371 X-Spam-Level: X-Spam-Status: No, score=-10.371 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-8, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xWbmJ0HxNgBs for ; Wed, 14 Aug 2013 18:42:13 -0700 (PDT) Received: from rcdn-iport-3.cisco.com (rcdn-iport-3.cisco.com [173.37.86.74]) by ietfa.amsl.com (Postfix) with ESMTP id 684A511E80E0 for ; Wed, 14 Aug 2013 18:42:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=12282; q=dns/txt; s=iport; t=1376530933; x=1377740533; h=from:to:cc:subject:date:message-id:in-reply-to: mime-version; bh=+iCs/5uZlP5Ss5LoXHHo78AxVepFq6ZZxJGmle1feL8=; b=C16silIpmH3Cnd7uVFyofI+oc+zhaJmditN7h13q+nGSf9Amfxd57LzE pt5cdbtVGEbvmmjq/Qt0UgNmtQNvldPTvD8V0O5CAOYKkvVfH0kswtxNC p2hr1yXgFmnYkLf9D5fCJbruiMrrJIiN9ErX4S33UN7YcadRG6IEzcVCr I=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Ag4FAIExDFKtJXHB/2dsb2JhbABbFoIsRDVQtkKHVYENgSEWdIIkAQEBAQMBAQEkRwsSAQgRAwECAQoCGygGAQoUCQgCBAENAQQIh3YDDwywPA2IXo1VgkogDQQHgxt3A4h1jChegxaKfoUngxuCKg X-IronPort-AV: E=Sophos;i="4.89,881,1367971200"; d="scan'208,217";a="247530993" Received: from rcdn-core2-6.cisco.com ([173.37.113.193]) by rcdn-iport-3.cisco.com with ESMTP; 15 Aug 2013 01:42:12 +0000 Received: from xhc-rcd-x04.cisco.com (xhc-rcd-x04.cisco.com [173.37.183.78]) by rcdn-core2-6.cisco.com (8.14.5/8.14.5) with ESMTP id r7F1gCAR018155 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 15 Aug 2013 01:42:12 GMT Received: from xmb-aln-x15.cisco.com ([169.254.9.134]) by xhc-rcd-x04.cisco.com ([fe80::200:5efe:173.37.183.34%12]) with mapi id 14.02.0318.004; Wed, 14 Aug 2013 20:42:12 -0500 From: "Rong Pan (ropan)" To: Dave Taht , Naeem Khademi Thread-Topic: [aqm] [iccrg] ECN support for PIE AQM in Linux Thread-Index: AQHOmRwtL/mswIODgUmaCmJqQZu7P5mViOyAgAA5LYD//5t4AA== Date: Thu, 15 Aug 2013 01:42:11 +0000 Message-ID: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.2.1.120420 x-originating-ip: [10.21.88.54] Content-Type: multipart/alternative; boundary="_000_BF2CBE9A994B89438F7F3B9FD58C2B7F26D79398xmbalnx15ciscoc_" MIME-Version: 1.0 Cc: "iccrg@irtf.org" , "aqm@ietf.org" , Vijay Subramanian , bloat Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 15 Aug 2013 01:42:19 -0000 --_000_BF2CBE9A994B89438F7F3B9FD58C2B7F26D79398xmbalnx15ciscoc_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable The byte mode being turned off is a pure implementation issue with Linux (w= hen the drop probability is very high, there are some rounding error issues= ). Vijay can probably articulate it better. The algorithm still has the byt= e mode operation. I think we can turn it back on for your test where there = is no extremely high drop rate and we are working on a fix, which will be a= vailable soon. Thanks, Rong From: Dave Taht > Date: Thursday, August 15, 2013 2:41 AM To: Naeem Khademi > Cc: "iccrg@irtf.org" >, "aqm@ietf.org" >, Vijay Subramanian >, bloat > Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux The pie patches that I was using until recently are here: http://snapon.lab.bufferbloat.net/~cero2/deb/patches/ Stephen hemminger and I cleaned up the original pie drop according to Kerne= lStyle in multiple ways. There is a rounding error problem in setting the t= arget value (we'd recomended they switch to ktime rather than jiffies) that I either ad= ded or was in there in the first place.... and we fixed a couple generic bu= gs... In addition to adding ecn support, we also added congestion notification to= (a la the linux red and fq_codel algorithms) to provide lossless feedback = into the host tcp stack. (this is, incidentally, 1 line of code) This optim= ization made it possible to do comparisons on host <-> host tests between t= hese qdiscs, with no router in-between. Syntactically the iproute2 patches could use identical syntax for time base= d values as the codel iproute2 patches. Haven't got around to that. (not my= job, really, to improve cisco's code) Now, I only now - today! - just noticed that the new pie code at cisco's si= te seems *substantially* different from the older pie code the above was ba= sed on, in that it no longer has an optimization for not dropping smaller p= ackets. (bytemode) This is going to invalidate most previous benchmarks of = pie that I know of, including all of mine, and most, if not all, of cablela= bs'. Sigh. I will try to refresh my patches and update the lab over the next week or s= o. On Wed, Aug 14, 2013 at 2:17 PM, Naeem Khademi > wrote: Hi Vijay This code implements basic ECN capability for PIE meaning that it would CE-mark packets that would be early-dropped otherwise. I have sanity-checked the code (e.g. successful ECN marking and receiver/sender reaction). We're now in process of conducting larger set of experiments with PIE w/wo ECN that will eventually be published as we move forward. Cheers, Naeem On Wed, Aug 14, 2013 at 8:27 PM, Vijay Subramanian > wrote: > Naeem, > > Thanks for testing and the contribution! We have ECN support in our curre= nt > code-under-development thanks to earlier help from Dave Taht but I will > merge your changes as appropriate and > submit it back for review. > > I will take a look at your code soon. Do you also have any test results > which you can share? > > Thanks again, > Vijay > > > On 14 August 2013 07:05, Naeem Khademi > wrote: >> >> Hi >> >> The already-existing PIE Linux code maintained at >> ftp://ftpeng.cisco.com/pie/ does not support ECN. Studying the interacti= on >> of ECN and AQM algos is probably of interest of some of us. I have added= ECN >> support to PIE Linux code and have given it some basic sanity tests. >> >> The modified code is available at >> http://heim.ifi.uio.no/naeemk/research/PIE/ >> >> I would appreciate any feedback. >> >> Regards, >> Naeem >> >> >> >> _______________________________________________ >> iccrg mailing list >> iccrg@irtf.org >> https://www.irtf.org/mailman/listinfo/iccrg >> > _______________________________________________ iccrg mailing list iccrg@irtf.org https://www.irtf.org/mailman/listinfo/iccrg -- Dave T=E4ht Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.= html --_000_BF2CBE9A994B89438F7F3B9FD58C2B7F26D79398xmbalnx15ciscoc_ Content-Type: text/html; charset="iso-8859-1" Content-ID: <8F5F5B0EBDE673449229AE8C9727AE75@emea.cisco.com> Content-Transfer-Encoding: quoted-printable

The byte mode being turned off is a pure implementation issue with Lin= ux (when the drop probability is very high, there are some rounding error i= ssues). Vijay can probably articulate it better. The algorithm still has th= e byte mode operation. I think we can turn it back on for your test where there is no extremely high drop ra= te and we are working on a fix, which will be available soon.

Thanks,

Rong


From: Dave Taht <dave.taht@gmail.com>
Date: Thursday, August 15, 2013 2:4= 1 AM
To: Naeem Khademi <naeem.khademi@gmail.com>
Cc: "iccrg@irtf.org" <icc= rg@irtf.org>, "aqm@ietf.org= " <aqm@ietf.org>, Vijay Subr= amanian <subramanian.vijay@gmail= .com>, bloat <bloa= t@lists.bufferbloat.net>
Subject: Re: [aqm] [iccrg] ECN supp= ort for PIE AQM in Linux

The pie patches that I was using until recently are here:<= br>
Stephen hemminger and I cleaned up the original pie drop according to = KernelStyle in multiple ways. There is a rounding error problem in setting = the target value
(we'd recomended they switch to ktime rather than jiffies) that I eith= er added or was in there in the first place.... and we fixed a couple gener= ic bugs...

In addition to adding ecn support, we also added congestion notificati= on to (a la the linux red and fq_codel algorithms) to provide lossless feed= back into the host tcp stack. (this is, incidentally, 1 line of code) This = optimization made it possible to do comparisons on host <-> host tests between these qdiscs, with no = router in-between.

Syntactically the iproute2 patches could use identical syntax for time= based values as the codel iproute2 patches. Haven't got around to that. (n= ot my job, really, to improve cisco's code)

Now, I only now - today! - just noticed that the new pie code at cisco= 's site seems *substantially* different from the older pie code the above w= as based on, in that it no longer has an optimization for not dropping smal= ler packets. (bytemode) This is going to invalidate most previous benchmarks of pie that I know of, includ= ing all of mine, and most, if not all, of cablelabs'. Sigh.

I will try to refresh my patches and update the lab over the next week= or so.



On Wed, Aug 14, 2013 at 2:17 PM, Naeem Khademi <= span dir=3D"ltr"> <naeem.khad= emi@gmail.com> wrote:
Hi Vijay

This code implements basic ECN capability for PIE meaning that it
would CE-mark packets that would be early-dropped otherwise. I have
sanity-checked the code (e.g. successful ECN marking and
receiver/sender reaction). We're now in process of conducting larger
set of experiments with PIE w/wo ECN that will eventually be published
as we move forward.

Cheers,
Naeem

On Wed, Aug 14, 2013 at 8:27 PM, Vijay Subramanian
<subramanian.vijay@gmail.= com> wrote:
> Naeem,
>
> Thanks for testing and the contribution! We have ECN support in our cu= rrent
> code-under-development thanks to earlier help from Dave Taht but I wil= l
> merge your changes as appropriate and
> submit it back for review.
>
> I will take a look at your code soon. Do you also have any test result= s
> which you can share?
>
> Thanks again,
> Vijay
>
>
> On 14 August 2013 07:05, Naeem Khademi <naeem.khademi@gmail.com> wrote:
>>
>> Hi
>>
>> The already-existing PIE Linux code maintained at
>> ftp://ft= peng.cisco.com/pie/ does not support ECN. Studying the interaction
>> of ECN and AQM algos is probably of interest of some of us. I have= added ECN
>> support to PIE Linux code and have given it some basic sanity test= s.
>>
>> The modified code is available at
>> http://heim.ifi.uio.no/naeemk/research/PIE/
>>
>> I would appreciate any feedback.
>>
>> Regards,
>> Naeem
>>
>>
>>
>> _______________________________________________
>> iccrg mailing list
>> iccrg@irtf.org
>> https://www.irtf.org/mailman/listinfo/iccrg
>>
>
_______________________________________________
iccrg mailing list
iccrg@irtf.org
h= ttps://www.irtf.org/mailman/listinfo/iccrg



--
Dave T=E4ht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html
--_000_BF2CBE9A994B89438F7F3B9FD58C2B7F26D79398xmbalnx15ciscoc_-- From subramanian.vijay@gmail.com Wed Aug 14 22:05:22 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6FF0111E8108 for ; Wed, 14 Aug 2013 22:05:22 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.372 X-Spam-Level: X-Spam-Status: No, score=-2.372 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JbgfbccGhFma for ; Wed, 14 Aug 2013 22:05:21 -0700 (PDT) Received: from mail-qe0-x233.google.com (mail-qe0-x233.google.com [IPv6:2607:f8b0:400d:c02::233]) by ietfa.amsl.com (Postfix) with ESMTP id 7D18721F90FD for ; Wed, 14 Aug 2013 22:05:21 -0700 (PDT) Received: by mail-qe0-f51.google.com with SMTP id nd7so199571qeb.38 for ; Wed, 14 Aug 2013 22:05:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=KzFi2nPv641TgcGQ9LnOzuijGu7PbeMoQRJ+o8SZrA4=; b=pTdlmLJWBxWYCcexaterA0HeuCG1wZA6E8gwaGBtw8DwG/PQdCKXkHmMHY3LfrnaKL oAdPZKAMZ8x8AC9lteVPiFR5aJXB2PJSM08CnNDDrXfD/nqHgikQaxcyeZB8RfVdHehV POwL/6vJtyeTbJ1d8dkBK0L3KbtlzXB+Ifr2rYuCpfscaxoR+peKj+2jhDJRtgQFdvW3 MzvS7jOjM3l8agg1T/t/rmuiviEkAjmHvIRI+lP0/GUmf6JtmkP9Q1s57zC+iY30d4Qb SaWQ8FwW0/0vjGvW+wGsnJw1DqYRYyry7kl96D5RzrMPVLovmSoTzOMk8tLkwv8YL6yR rCxA== MIME-Version: 1.0 X-Received: by 10.224.113.209 with SMTP id b17mr9338161qaq.85.1376543120934; Wed, 14 Aug 2013 22:05:20 -0700 (PDT) Received: by 10.49.105.232 with HTTP; Wed, 14 Aug 2013 22:05:20 -0700 (PDT) In-Reply-To: References: Date: Wed, 14 Aug 2013 22:05:20 -0700 Message-ID: From: Vijay Subramanian To: Dave Taht Content-Type: multipart/alternative; boundary=047d7bdcab90bdd22204e3f56bf7 Cc: "iccrg@irtf.org" , Naeem Khademi , bloat , "aqm@ietf.org" Subject: Re: [aqm] [iccrg] ECN support for PIE AQM in Linux X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 15 Aug 2013 05:05:22 -0000 --047d7bdcab90bdd22204e3f56bf7 Content-Type: text/plain; charset=ISO-8859-1 On 14 August 2013 17:41, Dave Taht wrote: > The pie patches that I was using until recently are here: > > http://snapon.lab.bufferbloat.net/~cero2/deb/patches/ > > Stephen hemminger and I cleaned up the original pie drop according to > KernelStyle in multiple ways. There is a rounding error problem in setting > the target value > (we'd recomended they switch to ktime rather than jiffies) that I either > added or was in there in the first place.... and we fixed a couple generic > bugs... > > > Dave, Thanks to you and Stephen for this cleanup. I have these changes in my queue. I had also promised Stephen that I would upstream the code to the linux networking list for review and inclusion and this is what I intend to do in the coming days. The idea was that once bufferbloat community has vetted the algorithm, we can make the needed implementation and style changes for kernel submission. I believe it is now time to submit it to the netdev list. i will merge the collective feedback received so far and send it out. Thanks, Vijay --047d7bdcab90bdd22204e3f56bf7 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable



On 14 August 2013 17:41, Dave Taht <dave.taht@gmail.com><= /span> wrote:
The pie patches that I was = using until recently are here:
Stephen hemminger and I cleaned up the original pie drop according to Kerne= lStyle in multiple ways. There is a rounding error problem in setting the t= arget value
(we'd recomended they switch to ktime rather= than jiffies) that I either added or was in there in the first place.... a= nd we fixed a couple generic bugs...


Dave,
Thanks to = you and Stephen for this cleanup. I have these changes in my queue. I had a= lso promised Stephen that I would upstream the code to the linux networking= list for review and inclusion and this is what I=A0
intend to do in the coming days. The idea was that once bufferbloat co= mmunity has vetted the algorithm, we can make the needed implementation and= style changes for kernel submission.=A0
I believe it is now time= to submit it to the netdev list. =A0i will merge the collective feedback r= eceived so far and send it out.

Thanks,
Vijay

=A0
--047d7bdcab90bdd22204e3f56bf7-- From rs@netapp.com Mon Aug 26 03:25:56 2013 Return-Path: X-Original-To: aqm@ietfa.amsl.com Delivered-To: aqm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6DF0B11E8183; Mon, 26 Aug 2013 03:25:55 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.429 X-Spam-Level: X-Spam-Status: No, score=-2.429 tagged_above=-999 required=5 tests=[AWL=-0.057, BAYES_00=-2.599, SARE_SUB_OBFU_Q1=0.227] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lKyc6k5nay86; Mon, 26 Aug 2013 03:25:50 -0700 (PDT) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id C792121F9B08; Mon, 26 Aug 2013 03:25:48 -0700 (PDT) X-IronPort-AV: E=Sophos;i="4.89,957,1367996400"; d="scan'208";a="44323219" Received: from vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) by mx11-out.netapp.com with ESMTP; 26 Aug 2013 03:25:48 -0700 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.122]) by vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) with mapi id 14.03.0123.003; Mon, 26 Aug 2013 03:25:37 -0700 From: "Scheffenegger, Richard" To: "aqm@ietf.org" , "iccrg@irtf.org" , "tsvwg@ietf.org" Thread-Topic: [aqm] Status after the IETF Thread-Index: AQHOlETyQn1xxH5YR0eYo6aA+VirqJmMUsUAgABBBGCAGs94sA== Date: Mon, 26 Aug 2013 10:25:37 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25CE081A@SACEXCMBX02-PRD.hq.netapp.com> References: <5203AD0F.5030907@neclab.eu> <20130808220232.GG3328@verdi> <012C3117EDDB3C4781FD802A8C27DD4F25C8D2ED@SACEXCMBX02-PRD.hq.netapp.com> In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25C8D2ED@SACEXCMBX02-PRD.hq.netapp.com> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.53] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: Wesley Eddy , Martin Stiemerling Subject: Re: [aqm] Status after the IETF X-BeenThere: aqm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: "Discussion list for active queue management and flow isolation." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 26 Aug 2013 10:25:56 -0000 Hi group, we, the BoF chairs, have asked for a tentative slot for AQM, assuming that = AQM will be chartered as a WG by the time of IETF88. Our responsible AD did= start the necessary preparations already! At this time, this is just a heads-up. To plan a proper slot, we'd like to = encourage early feedback. Best regards, Richard Scheffenegger