From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from EUR01-HE1-obe.outbound.protection.outlook.com (mail-he1eur01on0077.outbound.protection.outlook.com [104.47.0.77]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 045EB3CB35 for ; Wed, 18 Apr 2018 08:21:30 -0400 (EDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=darbyshire-bryant.me.uk; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version; bh=lZjcPq8tHX7Lz+REI82PhX+ERtLfNip9ZJVV/TgA/xo=; b=Pyr0bPh9cjjXSPsBfwjbl4aV4VnZEdt5ZdkQqKNwgNs6MMhxBngQcgHwsKdAbPuD1vQJt/0yc8gf0YqaeWjeborih/CYY6143pnm2svnTU02MHQY6VYiYVwigRr8INLZEXaLxrYfiQJWQV3ivvmtwIdCTu9+5RBjfy+RvkzpcbI= Received: from VI1PR0702MB3741.eurprd07.prod.outlook.com (52.134.3.159) by VI1PR0702MB3712.eurprd07.prod.outlook.com (52.134.3.154) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.696.7; Wed, 18 Apr 2018 12:21:28 +0000 Received: from VI1PR0702MB3741.eurprd07.prod.outlook.com ([fe80::3cf5:7237:f66:51b5]) by VI1PR0702MB3741.eurprd07.prod.outlook.com ([fe80::3cf5:7237:f66:51b5%2]) with mapi id 15.20.0696.013; Wed, 18 Apr 2018 12:21:27 +0000 From: Kevin Darbyshire-Bryant To: =?utf-8?B?VG9rZSBIw7hpbGFuZC1Kw7hyZ2Vuc2Vu?= CC: Jonathan Morton , "cake@lists.bufferbloat.net" Subject: Re: [Cake] A few puzzling Cake results Thread-Topic: [Cake] A few puzzling Cake results Thread-Index: AQHT1jB4sTDL6vBhakucJ8y9yVrRPaQE/UMAgAAEFwCAAWIvAIAAD6KA Date: Wed, 18 Apr 2018 12:21:27 +0000 Message-ID: <1B7176CA-41BC-4CF0-838D-871F0C858CF3@darbyshire-bryant.me.uk> References: <87vacq419h.fsf@toke.dk> <874lk9533l.fsf@toke.dk> <87604o3get.fsf@toke.dk> In-Reply-To: <87604o3get.fsf@toke.dk> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: authentication-results: toke.dk; dkim=none (message not signed) header.d=none;toke.dk; dmarc=none action=none header.from=darbyshire-bryant.me.uk; x-originating-ip: [2a02:c7f:1231:202b::dc83] x-ms-publictraffictype: Email x-microsoft-exchange-diagnostics: 1; VI1PR0702MB3712; 7:uLKHNvx4v3E58jnKj//xUt2yM50QLKfpS47sGXxb7POUXvVCalaBAg7U4qwv6McOBbuIRmGKzMcNttFQvrBxR6Ev5IzaXLNCmu+hju6ANqCg2/zWJyaF7il+CM9aZXWiaiAQA5GD5+7nQGwjk8okMLElLh/wNw9c0zax6GMWSw4gCoWZZXq0sNNbUT/fbpoeKUIKzI4sWQWylDb336qxDkyJ985nRhUD9Si0Ef5AFmtMI/DX/NaajnGn6SOqs+XI x-ms-exchange-antispam-srfa-diagnostics: SOS; x-microsoft-antispam: UriScan:; BCL:0; PCL:0; RULEID:(7020095)(4652020)(7021125)(5600026)(4534165)(7022125)(4603075)(4627221)(201702281549075)(7048125)(7024125)(7027125)(7028125)(7023125)(2017052603328)(7153060)(49563074)(7193020); SRVR:VI1PR0702MB3712; x-ms-traffictypediagnostic: VI1PR0702MB3712: x-microsoft-antispam-prvs: x-exchange-antispam-report-test: UriScan:(85827821059158)(226747003789276); x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(102415395)(6040522)(2401047)(5005006)(8121501046)(3231232)(944501327)(52105095)(3002001)(10201501046)(93006095)(93001095)(6041310)(20161123564045)(20161123560045)(2016111802025)(20161123562045)(20161123558120)(6072148)(6043046)(201708071742011); SRVR:VI1PR0702MB3712; BCL:0; PCL:0; RULEID:; SRVR:VI1PR0702MB3712; x-forefront-prvs: 06469BCC91 x-forefront-antispam-report: SFV:NSPM; SFS:(10009020)(376002)(346002)(396003)(366004)(39380400002)(39830400003)(2900100001)(97736004)(6116002)(46003)(39060400002)(53936002)(6246003)(6306002)(6512007)(33656002)(478600001)(5660300001)(81166006)(8676002)(966005)(305945005)(6436002)(25786009)(186003)(5250100002)(4326008)(74482002)(8936002)(82746002)(316002)(54906003)(446003)(7736002)(14454004)(11346002)(2616005)(99936001)(93886005)(53546011)(476003)(36756003)(6916009)(3660700001)(102836004)(76176011)(86362001)(83716003)(6506007)(2906002)(575784001)(6486002)(229853002)(99286004)(3280700002); DIR:OUT; SFP:1101; SCL:1; SRVR:VI1PR0702MB3712; H:VI1PR0702MB3741.eurprd07.prod.outlook.com; FPR:; SPF:None; LANG:en; MLV:ovrnspm; PTR:InfoNoRecords; x-microsoft-antispam-message-info: Y1DIvWoRKoZ8qOj7eMdTTj2LYjUI1dvizn8BSLTQfayISfL/Z8ptW7HbVXgDY0FL1GwhRhAlPy5QS9lwKeXxew9Pz7VGCgKurIKItI+oL3lDa02g673vaeP1JYGTH7k/HeWnkf7kXBikWpqAkrKXu965575Q9UcnvY33kgMX1U3rmGvZ/ngva83fj72a81D5 spamdiagnosticoutput: 1:99 spamdiagnosticmetadata: NSPM Content-Type: multipart/signed; boundary="Apple-Mail=_26D520C0-54F1-4690-8E6C-C3C74BF01335"; protocol="application/pgp-signature"; micalg=pgp-sha256 MIME-Version: 1.0 X-MS-Office365-Filtering-Correlation-Id: 824ac85d-af46-48c1-5b3e-08d5a526e904 X-OriginatorOrg: darbyshire-bryant.me.uk X-MS-Exchange-CrossTenant-Network-Message-Id: 824ac85d-af46-48c1-5b3e-08d5a526e904 X-MS-Exchange-CrossTenant-originalarrivaltime: 18 Apr 2018 12:21:27.8509 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: 9151708b-c553-406f-8e56-694f435154a4 X-MS-Exchange-Transport-CrossTenantHeadersStamped: VI1PR0702MB3712 X-List-Received-Date: Wed, 18 Apr 2018 12:21:31 -0000 --Apple-Mail=_26D520C0-54F1-4690-8E6C-C3C74BF01335 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8 > On 18 Apr 2018, at 12:25, Toke H=C3=B8iland-J=C3=B8rgensen = wrote: >=20 > Toke H=C3=B8iland-J=C3=B8rgensen writes: >=20 >> Jonathan Morton writes: >>=20 >>>> On 17 Apr, 2018, at 12:42 pm, Toke H=C3=B8iland-J=C3=B8rgensen = wrote: >>>>=20 >>>> - The TCP RTT of the 32 flows is *way* higher for Cake. FQ-CoDel >>>> controls TCP flow latency to around 65 ms, while for Cake it is all >>>> the way up around the 180ms mark. Is the Codel version in Cake too >>>> lenient, or what is going on here? >>>=20 >>> A recent change was to increase the target dynamically so that at >>> least 4 MTUs per flow could fit in each queue without AQM activity. >>> That should improve throughput in high-contention scenarios, but it >>> does come at the expense of intra-flow latency when it's relevant. >>=20 >> Ah, right, that might explain it. In the 128 flow case each flow has >> less than 100 Kbps available to it, so four MTUs are going to take a >> while to dequeue... >=20 > OK, so I went and looked at the code and found this: >=20 > bool over_target =3D sojourn > p->target && > sojourn > p->mtu_time * bulk_flows * 4; >=20 >=20 > Which means that we scale the allowed sojourn time for each flow by = the > time of four packets *times the number of bulk flows*. >=20 > So if there is one active bulk flow, we allow each flow to queue four > packets. But if there are ten active bulk flows, we allow *each* flow = to > queue *40* packets. >=20 > This completely breaks the isolation of different flows, and makes the > scaling of Cake *worse* than plain CoDel. >=20 > So why on earth would we do that? The thread that lead to that change: https://lists.bufferbloat.net/pipermail/cake/2017-December/003159.html Commits: 0d8f30faa3d4bb2bc87a382f18d8e0f3e4e56eac & the change to 4*bulk = flows 49776da5b93f03c8548e26f2d7982d553d1d226c Cheers, Kevin D-B 012C ACB2 28C6 C53E 9775 9123 B3A2 389B 9DE2 334A --Apple-Mail=_26D520C0-54F1-4690-8E6C-C3C74BF01335 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc Content-Description: Message signed with OpenPGP -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEASyssijGxT6XdZEjs6I4m53iM0oFAlrXOEcACgkQs6I4m53i M0rAWhAAnF9ZqgreZ3g6BRSDfyq2TiXFmJCQTI07fJwQOUPyq09IRNT5Fa8rOg7W MoWrEA9gk3l+JcA9z7NbUvnJLQbpYyjIi5U6uFgUgQuXFsk6HaJjg66EJ0GdD74B yoxhPuhxPzZ6hkYjXqxCsN91FtvAEXpXQ3SRNU8LpWyA63Y0E8w6s9JlC5Ztt4w4 Fc2wZNN019/Km7lKq9oAaBa7aafDUFVWED6yHNK9rqrV6/xTS07bug7GS5q55IsN z1Vpj2u1LRItqWY/BUWvwLQ4ZdwVtcMnqiQVW/N6KAnGoj6d24VtKZ6xTNfJwJBq z8jZ0ZePfRjEffJxZLq/XTJJfMV0IT/lFlUIB5BovEenyDI2yG0vhoXmrJaLhOrF 9mBBib+K7f9m79pyaKKx/Cus3rIJZyqDZxPX12dlPVvyh1CRqKjnvFOkrtK0aD86 fCQ7qX+1v4yeWvRbgGbWFUhTA2On5oWsx1fanuV7QpVnImwcpfU0duH3McV6AWh9 d7IRKZ8rSqXzqUmTajIiRzoODKQr20OvRUgfSLUYBrlOiKKzInLbYFEI+6KhDaXD 09dKhixslGGwIBRuuL7xaFEo3HPvph0Z4a781lbwUyx7ysxZnjJPw+XaVPwkerpv BPAHyaA4o0sCkr0yJAyMoMnYH0Bv1yOejkwiEdNwy0PD2gUK+Do= =kEVR -----END PGP SIGNATURE----- --Apple-Mail=_26D520C0-54F1-4690-8E6C-C3C74BF01335--