From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-oa1-x29.google.com (mail-oa1-x29.google.com [IPv6:2001:4860:4864:20::29]) (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 1BDD93B2A4 for ; Thu, 6 Oct 2022 04:06:49 -0400 (EDT) Received: by mail-oa1-x29.google.com with SMTP id 586e51a60fabf-1322fa1cf6fso1366182fac.6 for ; Thu, 06 Oct 2022 01:06:49 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=quora.org; s=google; h=to:subject:message-id:date:from:mime-version:from:to:cc:subject :date; bh=FqN1qnonFZwcmtey4fbqddlbJfOy+NkUaH6YqoyjaKc=; b=Z9BBShTNeIfp/zuuQMXa64w/y83HL/i0Ih523PZxo0u6zgHR8gwMrheJl55Q/SRaMk LT0O/pXwXIzVFBf0Ar5yACtQLAKkE3/IEk+s8KsTmV2MbrWcS7tHfA0S0L5G948N/2Zp Ru0gLIHsRVpNq7YNSBvdTPT3n9Ol9VRI5x3hg= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=to:subject:message-id:date:from:mime-version:x-gm-message-state :from:to:cc:subject:date; bh=FqN1qnonFZwcmtey4fbqddlbJfOy+NkUaH6YqoyjaKc=; b=GqxVk8fkkAfQ90E+COurj526pZyzmAbAtxEZfEAwS+xQ0q4uBZGAKNTk3SBKTtqFd4 k9YUnkqpZvHtYY0NtZhzKJmwHqMrhnoJtutSdJ9WQ7mXr5pDOTK+1iotgm5ieV/fEts+ q4b7LztrvdZTz3fa7sYOIDd3xQoGbl1ax7DC2Rj5H43DFeJLzCaR6HcZUALO6EFCmbps XYD6oGvogaPVQpS8qIaGBLbOlbWKCgAXA8axhc/kiG2J08CI8GjP2hlkiJcNiZwsGHDK 6FN//kGNKcD6SN+FnL/NcFJo21raDkTgQObtgB8y/dsqEbpSLwekIQCNyeVjWeCRkFK9 HwGA== X-Gm-Message-State: ACrzQf319cF4kOadPLVs/k8st3zrQTHaUg1DaXA1YK1l+b+uGg3XeBV+ T5lB1b+rTZS1mFEcW4J+oqWTVJL8b8PblJxAsSakERYEP6c= X-Google-Smtp-Source: AMsMyM752c75OKCrKoyoOcvdfK9CqfjX3q1iaGKdqpWg9YOSRDUiiT2tNiJMCuL9SoUXPhwwO+BLBBhMIB9OevJh63w= X-Received: by 2002:a05:6870:f289:b0:12c:22bf:1535 with SMTP id u9-20020a056870f28900b0012c22bf1535mr1956815oap.228.1665043608084; Thu, 06 Oct 2022 01:06:48 -0700 (PDT) MIME-Version: 1.0 From: Daniel J Blueman Date: Thu, 6 Oct 2022 16:06:37 +0800 Message-ID: To: cake@lists.bufferbloat.net Content-Type: text/plain; charset="UTF-8" Subject: [Cake] WiFI/shared media and bursting 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: Thu, 06 Oct 2022 08:06:49 -0000 Hi Dave, Toke et al, Given various challenges brought by WiFi (and related radio frequency systems): 1. the maximum throughput varies by media attenuation (eg range) and device capability, all else aside (ie isolated environment) 2. the upstream and downstream paths share the same media 3. the media is shared with other devices (increasing the noise floor) and networks cause 802.11 Clear Channel Assessment to defer transmission For these reasons any bandwidth estimation (autorate) may be valid only for 10s of milliseconds. Therefore for a given density of access point deployment, there should be a conservative throughput lower bound covering, for example 40Mbps (peak unidir would be 120Mbps). Since this is shared among upload and download, one could deploy CAKE with a fixed 20Mbps limit for upload and 20Mbps for download. A second approach might be to have a TBF or related scheme to allow per-stream bursts, limiting them to 15Mbps (or less) steady-state after a burst period. Though this defeats one aspect of CAKE, I'm using this approach to attempt to manage latency better. It's worth noting the Linux drivers I'm using aren't ath10k or such, so do over-buffer. Does anyone have any advice or experience on this or another scheme with WiFi or related? Thanks, Dan -- Daniel J Blueman