From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-x22c.google.com (mail-qk0-x22c.google.com [IPv6:2607:f8b0:400d:c09::22c]) (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 9E0343B29E for ; Sun, 29 Oct 2017 18:39:48 -0400 (EDT) Received: by mail-qk0-x22c.google.com with SMTP id q83so14070498qke.6 for ; Sun, 29 Oct 2017 15:39:48 -0700 (PDT) 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=HeZPmB8rAl95sldcgObtvigluqMo6dcaW9g2SAtF+ko=; b=mBG84/15YRu/eKr8ZZQR7/BEa6D7ybKtJs3bmrqcFpGpGyD/a3A9AfMHEbJt02vkoY dxWiucTGvDK327skGqXO2iJUUiJC4GOiva5u2i/is+JlU8FAM/xUCqoltNHOmJ/6/zY8 X5GSiUWPkqY6hOsWZ41n2vm0wm/mkOvGfs7Kc41wR/6EKHWjAL2wfD9cnKuvpvYBNIT6 f1w8unwCUyLUKlrDWdZvlZ7aiP2S+f4cMno8YSnIjRiBlX6wa+ahprdeA4xGc9FWRwN9 UsM++XFI1wSPPjiUsB6y9VDYPkJ29OHORURCxh7evtdOrzfK+VFoANHMyjb2ncHMsIL4 V5qA== 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=HeZPmB8rAl95sldcgObtvigluqMo6dcaW9g2SAtF+ko=; b=XX8gPOFNTHe8MhEBmgUPGu3N/VSUMlNTGzm4sbtLf+S0O0LQf7vGkllMpSQJbXjHPD Kn+9MgaDknT0WJ0lscXjThrkbolE5g5zjpfB4Ebkh3TGtzh1vFxLU6MYth4AeQ0VixC1 U1y5eCBsUusK+nlyvDUUm05to1UraibelzHSIdaZFsQVTxfNc1xj5WN5c0ecSOcX7nkX SUUuwHHVMS9TR5cFJYozcFKl8vmVKh9PFTo+cAH6rJ+Xz/kn1tTZlHdKOK6JHIM54zzJ iiUYTe434Xd7KVZBD3VRViTuVM4TWg4hVRy3sXPFLO/Ho6unbn00aGafEIp8LLwVwRGU Nqvg== X-Gm-Message-State: AMCzsaXAsrMICBZtvK6fOAsZnc+efbGpDT3rxnAixOWhF/jZGkQ95tJz +2jlSAFrrGZysxCI3ixcUoHRNwOh8o3OM08StLM= X-Google-Smtp-Source: ABhQp+ThfYkxeYsJEuatWJGaGeSSCa+AMuG5ogi49B9rR0UT4xvCKeajWPsqHPCRt90K8hP0eJbnZ5qD86NciiFvtrs= X-Received: by 10.55.162.198 with SMTP id l189mr10707096qke.168.1509316788232; Sun, 29 Oct 2017 15:39:48 -0700 (PDT) MIME-Version: 1.0 Received: by 10.140.80.133 with HTTP; Sun, 29 Oct 2017 15:39:47 -0700 (PDT) Received: by 10.140.80.133 with HTTP; Sun, 29 Oct 2017 15:39:47 -0700 (PDT) In-Reply-To: References: From: Jonathan Morton Date: Mon, 30 Oct 2017 00:39:47 +0200 Message-ID: To: Dave Taht Cc: Adrian Popescu , Cake List Content-Type: multipart/alternative; boundary="001a114d6dbc042b8d055cb73320" Subject: Re: [Cake] CAKE set-associative hashing 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: Sun, 29 Oct 2017 22:39:48 -0000 --001a114d6dbc042b8d055cb73320 Content-Type: text/plain; charset="UTF-8" In CPUs, 4-way is now common in first-level caches, 16-way at last-level. It is a tradeoff between guaranteed hit latency (necessary in synchronous hardware) and collision rate. In software, average hit latency is usually more relevant, and worst-case throughput otherwise. IMHO, having a separate qdisc tree per hardware queue is the wrong design. Better would be a single qdisc (or tree thereof) that can be drained concurrently into all queues. That currently seems to be easiest to arrange by putting the qdiscs in userspace, but theoretically that need not be so. - Jonathan Morton --001a114d6dbc042b8d055cb73320 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

In CPUs, 4-way is now common in first-level caches, 16-way a= t last-level.=C2=A0 It is a tradeoff between guaranteed hit latency (necess= ary in synchronous hardware) and collision rate.=C2=A0 In software, average= hit latency is usually more relevant, and worst-case throughput otherwise.=

IMHO, having a separate qdisc tree per hardware queue is the= wrong design.=C2=A0 Better would be a single qdisc (or tree thereof) that = can be drained concurrently into all queues.=C2=A0 That currently seems to = be easiest to arrange by putting the qdiscs in userspace, but theoretically= that need not be so.

- Jonathan Morton

--001a114d6dbc042b8d055cb73320--