From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-oi1-x231.google.com (mail-oi1-x231.google.com [IPv6:2607:f8b0:4864:20::231]) (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 5AF753B2A4 for ; Tue, 3 Jan 2023 11:41:48 -0500 (EST) Received: by mail-oi1-x231.google.com with SMTP id d127so26156090oif.12 for ; Tue, 03 Jan 2023 08:41:48 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=domos-no.20210112.gappssmtp.com; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=utljfRR02ti4zL0tkg8zKhS+kULdKTggonfSPPf2+z4=; b=zE3hyjlluT3lcAYKlqI17FSi5B6hLwrDCCB/zR1O9FRn/F8bbu3+xCXs9mj5SeFoC/ akCDYbBzYWUTI3VP545oaJAMZhQ026ejuX8lOzzfiHde1l3/wW/0k/K6ViHGeTaaR+Ql ksXIF5wDHjRObm66MtoRRpcL2TNhOWTDh71KExEXTCw3FtHR9NKX0QbqcV9RGNx1YYwF M+hila74Aa0dmUV56REx0hbqLh0b4/0NEvDCb+WoEoaCfMZDOjBysfJNiuP7FMTNskTL Ac6n1e4ZWAKuR5WmblB3Lh3pyI5CEes00rMlBXLDVra7l1azeEUgERILizWeZtIsbdj1 mWsQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=utljfRR02ti4zL0tkg8zKhS+kULdKTggonfSPPf2+z4=; b=RgSknY7IGJx3tPiAP5fTIjwK/GKDUcfJLivGIQliDqSpgxdT10NXZ6GaoJAVTqekYW 2HX5PY8WLUa5jYqTf5E+Lq05zzVU4B7mIAk4QHUzdKv9ElqhILmOn4LiIQxRK0yVvhQj 82ZZuQftqRToqEZk7aH1UJQAz9oqBgwJbTibzAuzcBdQ897GU/zGYDCkRZLQXvTKX7Iz AAzYXtUo13BCvFvBBzBvPjmZ3uI7PYPZO+pPPklf0LdZCWzawN3mTJdyR3y0VhPTu8Yb ZKgJDEvS+GDmDoMGI248sDULNi33GVJR8GNmwR7R8uAxgofQLvpdhmRn3tVDEwBcvL99 cFKw== X-Gm-Message-State: AFqh2kroZysEFvY+XHKakMTza+d2KXCq6TG702Wo6Wfz1KzvrzQMuzlO QkXvz1dlBOC1ZkZH2PoO6hCqMVBxWJ38p4OyxmxdJrrINUj9Vw== X-Google-Smtp-Source: AMrXdXseDDQYfIaxV36ENa2RvL7viuUv1nnadvWh40UhgrfWPbfl5rfcWsCXtAI5P/G8RG7nExEPBLtlTrT+RdXuWO8= X-Received: by 2002:a05:6808:9b6:b0:363:a03b:6cdd with SMTP id e22-20020a05680809b600b00363a03b6cddmr674426oig.104.1672764107475; Tue, 03 Jan 2023 08:41:47 -0800 (PST) MIME-Version: 1.0 References: <27202.1672758483@localhost> In-Reply-To: <27202.1672758483@localhost> From: =?UTF-8?Q?Bj=C3=B8rn_Ivar_Teigen?= Date: Tue, 3 Jan 2023 17:41:36 +0100 Message-ID: To: Michael Richardson Cc: bloat Content-Type: multipart/alternative; boundary="0000000000006b611905f15ebecf" Subject: Re: [Bloat] PhD thesis with results related to buffering needs on variable-capacity links X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 03 Jan 2023 16:41:48 -0000 --0000000000006b611905f15ebecf Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Michael, On Tue, 3 Jan 2023 at 16:08, Michael Richardson wrote: > > Thank you for sharing your thesis! > > Bj=C3=B8rn Ivar Teigen via Bloat wrote: > > The thesis begins by investigating which performance issues are mos= t > > prevalent in today's WiFi networks. We show that both queuing laten= cy > > and the WiFi protocol specification itself are significant > > contributors. By building a model of the WiFi protocol behavior we > > quantify the performance of the protocol in terms of quality > > attenuation. We find that significant performance variability is an > > inherent consequence of the protocol design. > > I guess that your thesis is mostly technical. > (I haven't clicked on the link yet) > You are right. It is focused on analyzing models of the WiFi MAC layer to calculate how good (or bad!) latency and packet loss gets in various scenarios, and how that latency and packet loss in turn affects congestion control algorithms. > > I wonder if there is work that might occur from the business department e= nd > of things. Why do we have 15 years of WiFi optimization, which seem to b= e > taking us further away from low latency. Are the new protocols actually > improving the situation for the end consumer? My impression is no. > Nobody notices because the quality is so unpredictable. > Through my work in Domos I have at least some insight into this. My impression is that it starts with business decisions, where the focus has been on maximum throughput numbers because that's what consumers think they want. It also takes effort, focus, and (most importantly) money to remove bufferbloat, so unless there is sufficient incentives from the commercial side of things it doesn't happen. I do think there is increasing awareness of latency (under load) as an important factor for user experience. We certainly see that from some of the (dare I say thought-leading) ISPs we're working with. I think there's hope we will see more good low-latency solutions deployed in the near future. Cheers, --=20 Bj=C3=B8rn Ivar Teigen Head of Research +47 47335952 | bjorn@domos.no | www.domos.no --0000000000006b611905f15ebecf Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Hi Michael,

<= div dir=3D"ltr" class=3D"gmail_attr">On Tue, 3 Jan 2023 at 16:08, Michael R= ichardson <mcr@sandelman.ca> = wrote:

Thank you for sharing your thesis!

Bj=C3=B8rn Ivar Teigen via Bloat wrote:
=C2=A0 =C2=A0 > The thesis begins by investigating which performance iss= ues are most
=C2=A0 =C2=A0 > prevalent in today's WiFi networks. We show that bot= h queuing latency
=C2=A0 =C2=A0 > and the WiFi protocol specification itself are significa= nt
=C2=A0 =C2=A0 > contributors. By building a model of the WiFi protocol b= ehavior we
=C2=A0 =C2=A0 > quantify the performance of the protocol in terms of qua= lity
=C2=A0 =C2=A0 > attenuation. We find that significant performance variab= ility is an
=C2=A0 =C2=A0 > inherent consequence of the protocol design.

I guess that your thesis is mostly technical.
(I haven't clicked on the link yet)
You are right.= It is focused on analyzing models of the WiFi MAC layer to calculate how g= ood (or bad!) latency and packet loss gets in various scenarios, and how th= at latency and packet loss in turn affects congestion control algorithms.

I wonder if there is work that might occur from the business department end=
of things.=C2=A0 Why do we have 15 years of WiFi optimization, which seem t= o be
taking us further away from low latency.=C2=A0 Are the new protocols actual= ly
improving the situation for the end consumer?=C2=A0 My impression is no. Nobody notices because the quality is so unpredictable.

Through my work in Domos I have at least some insight into= this. My impression is that it starts with business decisions, where the f= ocus has been on maximum throughput numbers because that's what consume= rs think they want. It also takes effort, focus, and (most importantly) mon= ey to remove bufferbloat, so unless there is sufficient incentives from the= commercial side of things it doesn't happen.
I do think ther= e is increasing awareness of latency (under load) as an important factor fo= r user experience. We certainly see that from some of the (dare I say thoug= ht-leading) ISPs we're working with.

I think t= here's hope we will see more good low-latency solutions deployed in the= near future.

Cheers,
--
Bj=C3=B8rn Iva= r Teigen
Head of Re= search
+47 = 47335952 | bjorn@domos.no= =C2=A0|<= span>=C2=A0www.domos.no
--0000000000006b611905f15ebecf--