From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp126.iad3a.emailsrvr.com (smtp126.iad3a.emailsrvr.com [173.203.187.126]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 2EFF33B29E for ; Wed, 10 Oct 2018 17:53:48 -0400 (EDT) Received: from smtp24.relay.iad3a.emailsrvr.com (localhost [127.0.0.1]) by smtp24.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id 00C3A24BF8; Wed, 10 Oct 2018 17:53:47 -0400 (EDT) X-SMTPDoctor-Processed: csmtpprox beta Received: from smtp24.relay.iad3a.emailsrvr.com (localhost [127.0.0.1]) by smtp24.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id EFDAC24CB5; Wed, 10 Oct 2018 17:53:47 -0400 (EDT) Received: from app46.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by smtp24.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id D217924C2B; Wed, 10 Oct 2018 17:53:47 -0400 (EDT) X-Sender-Id: dpreed@deepplum.com Received: from app46.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by 0.0.0.0:25 (trex/5.7.12); Wed, 10 Oct 2018 17:53:47 -0400 Received: from deepplum.com (localhost.localdomain [127.0.0.1]) by app46.wa-webapps.iad3a (Postfix) with ESMTP id BE3384008C; Wed, 10 Oct 2018 17:53:47 -0400 (EDT) Received: by apps.rackspace.com (Authenticated sender: dpreed@deepplum.com, from: dpreed@deepplum.com) with HTTP; Wed, 10 Oct 2018 17:53:47 -0400 (EDT) X-Auth-ID: dpreed@deepplum.com Date: Wed, 10 Oct 2018 17:53:47 -0400 (EDT) From: "David P. Reed" To: "Dave Taht" Cc: "Make-Wifi-fast" MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_20181010175347000000_29678" Importance: Normal X-Priority: 3 (Normal) X-Type: html In-Reply-To: References: Message-ID: <1539208427.776210290@apps.rackspace.com> X-Mailer: webmail/15.4.3-RC Subject: Re: [Make-wifi-fast] sometimes I get really tired of the wifi mac X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 10 Oct 2018 21:53:48 -0000 ------=_20181010175347000000_29678 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable =0AUmm... given that the WiFi MAC is pretty poor with 1500 byte packets in = today's 5 GHz operation, how badly will it work in WiGig?=0A =0AI'd also no= te that all that bandwidth is great for serial high speed links, but feedin= g it and handling it when you are transporting typical TCP/IP packets requi= res aggregating a lot of flows, and thus a pretty hefty amount of fan-out/f= an-in multiplexing.=0A =0AAgain, people obsess about throughput, but aren't= considering latency.=0A =0AWe'll probably see stuff produced that has the = "feature" of huge buffers, with bufferbloat causing much of the capacity to= be wasted.=0AOr, instead of demand-driven packet switching, just dividing = up the capacity into fixed rate "channels" at low rates, like telephone cal= ls.=0A =0AYou let the hardware guys focus on the pipes, and the protocols g= o to shit.=0A =0A-----Original Message-----=0AFrom: "Dave Taht" =0ASent: Wednesday, October 10, 2018 3:04pm=0ATo: "Make-Wifi-fast= " =0ASubject: [Make-wifi-fast] someti= mes I get really tired of the wifi mac=0A=0A=0A=0Ahttps://code.fb.com/conne= ctivity/introducing-facebook-s-new-terrestrial-connectivity-systems-terragr= aph-and-project-aries/=0A=0A-- =0A=0ADave T=C3=A4ht=0ACTO, TekLibre, LLC=0A= http://www.teklibre.com=0ATel: 1-831-205-9740=0A___________________________= ____________________=0AMake-wifi-fast mailing list=0AMake-wifi-fast@lists.b= ufferbloat.net=0Ahttps://lists.bufferbloat.net/listinfo/make-wifi-fast=0A--= =0AReed Online Ltd is a company registered in England and Wales. Company = =0ARegistration Number: 6317279.Registered Office: Academy Court, 94 Chance= ry =0ALane, London WC2A 1DT.=0A=0A ------=_20181010175347000000_29678 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Umm... given that the = WiFi MAC is pretty poor with 1500 byte packets in today's 5 GHz operation, = how badly will it work in WiGig?

=0A

 

=0AI'd also note that all that bandwidth is great for seria= l high speed links, but feeding it and handling it when you are transportin= g typical TCP/IP packets requires aggregating a lot of flows, and thus a pr= etty hefty amount of fan-out/fan-in multiplexing.

=0A

Again, people obsess about throughput, = but aren't considering latency.

=0A

 

=0A

We'll probably see stuff produced that has the "feature" = of huge buffers, with bufferbloat causing much of the capacity to be wasted= .

=0A

Or, instead of demand-driven packet switching,= just dividing up the capacity into fixed rate "channels" at low rates, lik= e telephone calls.

=0A

 

=0A

You let the hardware guys focus on the pipes, and the protocols go to = shit.

=0A

 

=0A

-----Ori= ginal Message-----
From: "Dave Taht" <dave.taht@gmail.com>
= Sent: Wednesday, October 10, 2018 3:04pm
To: "Make-Wifi-fast" <make= -wifi-fast@lists.bufferbloat.net>
Subject: [Make-wifi-fast] sometim= es I get really tired of the wifi mac

=0A
=0A

https://code.fb.com/connectivity/intr= oducing-facebook-s-new-terrestrial-connectivity-systems-terragraph-and-proj= ect-aries/

--

Dave T=C3=A4ht
CTO, TekLibre, LLC=
http://www.teklibre.com
Tel: 1-831-205-9740
_______________= ________________________________
Make-wifi-fast mailing list
Make= -wifi-fast@lists.bufferbloat.net
https://lists.bufferbloat.net/listinf= o/make-wifi-fast
--
Reed Online Ltd is a company registered in E= ngland and Wales. Company
Registration Number: 6317279.Registere= d Office: Academy Court, 94 Chancery
Lane, London WC2A 1DT.

=0A
 ------=_20181010175347000000_29678--