From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lb0-x22a.google.com (mail-lb0-x22a.google.com [IPv6:2a00:1450:4010:c04::22a]) by lists.bufferbloat.net (Postfix) with ESMTPS id 3CC9B3ED8F for ; Mon, 21 Dec 2015 19:30:15 -0500 (EST) Received: by mail-lb0-x22a.google.com with SMTP id sv6so16227696lbb.0 for ; Mon, 21 Dec 2015 16:30:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=xqZFaNmcnbrzvxIREWWl8qxKU085fUj3hAoN1lmhoog=; b=iGhB17niv6mXeBpkAPCBQn7MTvqIWsjawh23nILsSh3UwlLOc8uVvrZQ+EgJikWldk 4NJtHB2Y7Uo9D6gYrM8aTxwbuGp2J1JwFc/mRAqXZ7niXglzqJjDo5psUO4Fz78+BOuS l9FRfUC4fATviGEqxnZhnV4ODHQUH9EPCYiPDOxD0ek9EyAMZUmuNtaKNITbzzlyK3NE QB8BA1SXmmJJd8vul2LkKPJy1ITrhU1BHsf55O9blHStJtDZmlrbCdhqpeXqm0VxVOrc BG3NUuWPDftYY1E2xuZZMyQMg4KXo5fBDf6p35KZ9UhQFL2MPVbRyObVTOeUyAserxqQ Cedw== X-Received: by 10.112.45.138 with SMTP id n10mr94711lbm.100.1450744213651; Mon, 21 Dec 2015 16:30:13 -0800 (PST) Received: from bass.home.chromatix.fi (37-33-99-74.bb.dnainternet.fi. [37.33.99.74]) by smtp.gmail.com with ESMTPSA id p1sm2932108lbo.30.2015.12.21.16.30.12 (version=TLS1 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Mon, 21 Dec 2015 16:30:13 -0800 (PST) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 9.2 \(3112\)) From: Jonathan Morton In-Reply-To: Date: Tue, 22 Dec 2015 02:30:10 +0200 Cc: =?utf-8?Q?Bj=C3=B6rn_Gr=C3=B6nvall?= , cake@lists.bufferbloat.net Content-Transfer-Encoding: quoted-printable Message-Id: <5DB7FF46-9A7B-497C-A0B8-BDF84E067253@gmail.com> References: To: Dave Taht X-Mailer: Apple Mail (2.3112) Subject: Re: [Cake] and the bad cpu news on arm is X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 22 Dec 2015 00:30:15 -0000 > On 20 Dec, 2015, at 14:59, Dave Taht wrote: >=20 > I do happen to like the set associativity idea, but it's proving > really hard to show its value with any existing tests we have. I=E2=80=99m sure we had those sorts of tests earlier on. The 50:1 test = might have been one of them. Certainly, measuring Jain=E2=80=99s = Fairness Index across the 50 flows should be revealing; with perfect = flow isolation it should be close to unity, while with uncorrected hash = collisions it should be noticeably inferior. It=E2=80=99s certainly easy to detect when it=E2=80=99s avoiding hash = collisions through the stats output, as that=E2=80=99s when the = =E2=80=9Cindirect hit=E2=80=9D counters start incrementing. Perhaps when the next stage of flow isolation is actually in and = working, the benefits will be easier to measure directly on a human = scale. I don=E2=80=99t think I=E2=80=99d be confident in my current = approach to the problem unless the set-associative hash was in place. - Jonathan Morton