From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-x243.google.com (mail-qk0-x243.google.com [IPv6:2607:f8b0:400d:c09::243]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 082E33B2A4 for ; Mon, 27 Nov 2017 13:28:14 -0500 (EST) Received: by mail-qk0-x243.google.com with SMTP id p19so33687887qke.2 for ; Mon, 27 Nov 2017 10:28:14 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=WTZpagimm1btCnzbzoySpBg4l6L3PJGBcsbAwzHlyqM=; b=lidAZWbRfkBc5iIicZbYJeuGpme6q+na3hqAEKHgQqs3qxllqTDBFj/PpcI/OgJBl5 yCF7yYjb8xRFeDOxZNvtrmNdI+niqkFZehDb/78vy2N2xT+ujes3UPjQvbmjiLZU1ym3 hHM8MU02LUTJ9Nb6Y6+AlAxXt0Bav5mXtSpsX82UXTs950RbWqUknHG7TzYPAPGAFhwP jWvTOmibVKI/PrJLWs4p9hchlcsj+03ampCSHb8p0tq0FIYo3q1RCHTy64ZD5TPC4Tdg 11aBLF0f5hyNRFVuk2/j19W71bvcl4NaZmTUzAQWCi0AT8mB3UgOuKXYjN9IWjb4pYSR 2QzQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=WTZpagimm1btCnzbzoySpBg4l6L3PJGBcsbAwzHlyqM=; b=XekNi3EAQiYkoORSoh16MVdOn9QAtHkZYuoZUavIqdHTDYLvmXXQVW42nZeCtkNsfw VuXFJZj7OPph0CiF9mSWpeTLm5p0yXspJirc7RxRPaOETAmLk2x29FyFI/2D4j7yMGnn 6nexZ0OtDHcpb0e26xGxaN1Qd0MDZre2mkmDM/oFkcY/3gXeY5TtFm9uHaSkoxqQPwnw BuMo7at1PbzsBh8p8iiJkb58Vc5Ki4MbyRSZuXect1zzwOvm+H9RXe3JiZpXC2gw+5S+ RSkYHnffvDf10zIBBwQVEGadORRYSAaF9AO31F1tRLvvNfxNkoiwfqSw+Y6hxCrhWXNi G9IA== X-Gm-Message-State: AJaThX7Ba0S4AQR+JNF9QZHdehUUJcw5iarPKvMDq/nhduqXactrKKvW oT3WX22eLzR9wgIQHmFhpD3EaWNZ4po1j8OtSUI= X-Google-Smtp-Source: AGs4zMZikWBOIUIgacVgXDEBN4yYiOURT7sV/kqj/YnhPf/91aone8uXyTUkbv8/CDb9YKr7kI7qTh/gFEUpaUKMCVU= X-Received: by 10.55.121.67 with SMTP id u64mr39058475qkc.317.1511807293343; Mon, 27 Nov 2017 10:28:13 -0800 (PST) MIME-Version: 1.0 Received: by 10.140.102.179 with HTTP; Mon, 27 Nov 2017 10:28:12 -0800 (PST) Received: by 10.140.102.179 with HTTP; Mon, 27 Nov 2017 10:28:12 -0800 (PST) In-Reply-To: <4939D5F5-B880-424B-874A-477E75B0D0A1@gmail.com> References: <7B914FA7-38A6-4113-9B2C-D6246873676A@gmail.com> <1493791193941.83458@telenor.com> <53AFD81B-F9F5-4DE1-A418-A177D0F339E8@gmail.com> <5FD7BE71-CD6B-4C2F-8149-54763F43C519@gmx.de> <4939D5F5-B880-424B-874A-477E75B0D0A1@gmail.com> From: Jonathan Morton Date: Mon, 27 Nov 2017 20:28:12 +0200 Message-ID: To: Pete Heist Cc: Sebastian Moeller , Cake List Content-Type: multipart/alternative; boundary="94eb2c0578d0b04ca9055efb10a0" Subject: Re: [Cake] Recomended HW to run cake and fq_codel? 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: Mon, 27 Nov 2017 18:28:14 -0000 --94eb2c0578d0b04ca9055efb10a0 Content-Type: text/plain; charset="UTF-8" An important factor when designing the test is the difference between intra-flow and inter-flow induced latencies, as well as the baseline latency. In general, AQM by itself controls intra-flow induced latency, while flow isolation (commonly FQ) controls inter-flow induced latency. I consider the latter to be more important to measure. Baseline latency is a factor of the underlying network topology, and is the type of latency most often measured. It should be measured in the no-load condition, but the choice of remote endpoint is critical. Large ISPs could gain an unfair advantage if they can provide a qualifying endpoint within their network, closer to the last mile links than most realistic Internet services. Conversely, ISPs are unlikely to endorse a measurement scheme which places the endpoints too far away from them. One reasonable possibility is to use DNS lookups to randomly-selected gTLDs as the benchmark. There are gTLD DNS servers well-placed in essentially all regions of interest, and effective DNS caching is a legitimate means for an ISP to improve their customers' internet performance. Random lookups (especially of domains which are known to not exist) should defeat the effects of such caching. Induced latency can then be measured by applying a load and comparing the new latency measurement to the baseline. This load can simultaneously be used to measure available throughput. The tests on dslreports offer a decent example of how to do this, but it would be necessary to standardise the load. - Jonathan Morton --94eb2c0578d0b04ca9055efb10a0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

An important factor when designing the test is the differenc= e between intra-flow and inter-flow induced latencies, as well as the basel= ine latency.

In general, AQM by itself controls intra-flow induced latenc= y, while flow isolation (commonly FQ) controls inter-flow induced latency.= =C2=A0 I consider the latter to be more important to measure.

Baseline latency is a factor of the underlying network topol= ogy, and is the type of latency most often measured.=C2=A0 It should be mea= sured in the no-load condition, but the choice of remote endpoint is critic= al.=C2=A0 Large ISPs could gain an unfair advantage if they can provide a q= ualifying endpoint within their network, closer to the last mile links than= most realistic Internet services.=C2=A0 Conversely, ISPs are unlikely to e= ndorse a measurement scheme which places the endpoints too far away from th= em.

One reasonable possibility is to use DNS lookups to randomly= -selected gTLDs as the benchmark.=C2=A0 There are gTLD DNS servers well-pla= ced in essentially all regions of interest, and effective DNS caching is a = legitimate means for an ISP to improve their customers' internet perfor= mance.=C2=A0 Random lookups (especially of domains which are known to not e= xist) should defeat the effects of such caching.

Induced latency can then be measured by applying a load and = comparing the new latency measurement to the baseline.=C2=A0 This load can = simultaneously be used to measure available throughput.=C2=A0 The tests on = dslreports offer a decent example of how to do this, but it would be necess= ary to standardise the load.

- Jonathan Morton

--94eb2c0578d0b04ca9055efb10a0--