From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ob0-x230.google.com (mail-ob0-x230.google.com [IPv6:2607:f8b0:4003:c01::230]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 680FB21F19B for ; Mon, 15 Jun 2015 15:52:00 -0700 (PDT) Received: by obctg8 with SMTP id tg8so15198745obc.3 for ; Mon, 15 Jun 2015 15:52:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type :content-transfer-encoding; bh=nfKXEr/WAoJyxe8PyVyhvc3CyvHingo8Ne5GwQamjQo=; b=QPk+O21nF9TtSda/VMioicy6+zvYRr3YvmZAsvsTordR2f8rLa2AkGKbWgrLB5Gyqa G0Y8fQHVAz52AExCC7RClBBFMA/KZDu379DH87OUFgc54utyWKF8RaJ5mgXEQzMB5Na4 xFenkq92vev1NLoawfPcKz+IQpVYDqOtmZAm+1BbcK0jMOewEuh9wAzC1EMi1xTAGmFz Evcx5hCw4GZysVHIu7UdgJoCbu2J1KE/HQRBCESH0PS8YxDzD7fy0Yozy3RLH4F9fptY Vv9+PeeB0MI+Pj+Rmti+nFzhLkbVkwUhfeMMF3tseRi5JvLytI4Iy5/l7ShYnL5BsA+5 fm1Q== MIME-Version: 1.0 X-Received: by 10.202.227.15 with SMTP id a15mr24893968oih.59.1434408720290; Mon, 15 Jun 2015 15:52:00 -0700 (PDT) Received: by 10.202.105.129 with HTTP; Mon, 15 Jun 2015 15:52:00 -0700 (PDT) Date: Mon, 15 Jun 2015 15:52:00 -0700 Message-ID: From: Dave Taht To: cake@lists.bufferbloat.net Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: [Cake] fragmenting for lowest per flow latency X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.13 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, 15 Jun 2015 22:52:29 -0000 (I should probably create an alias for the crazy ideas like "devnull") but I have kind of wondered just how bad the fragmentation problem really w= as, and with peeling, already, we could also add fragmentation support, to create tons of fragments... and see what breaks, and get even lower per small packet flow latency when larger packets were present. I actually have a use case for this personally in that dnssec's rules (eDNS0) for fragmentation were a real pain in the arse of late, and *knowing* devices we work on handle fragments appropriately would be nice. --=20 Dave T=C3=A4ht What will it take to vastly improve wifi for everyone? https://plus.google.com/u/0/explore/makewififast