From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lf1-x12f.google.com (mail-lf1-x12f.google.com [IPv6:2a00:1450:4864:20::12f]) (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 613773B2A4 for ; Mon, 14 Feb 2022 11:32:57 -0500 (EST) Received: by mail-lf1-x12f.google.com with SMTP id j7so14372591lfu.6 for ; Mon, 14 Feb 2022 08:32:57 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=domos-no.20210112.gappssmtp.com; s=20210112; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=ECKUt25P3rRJAm43wEfY7GKrEVvZnxb1Bva2kHWnE98=; b=TLBHNOT55gDLh8+l4kyjjJ5yMOseG6kR/Ep+mZr/EQz8bH8K+oVH3nYpnFMY1n18B8 mjeijjaMzKiD9QUOaXpKmyQOkgd0JX+khOgtcjXxom6WcYede8l1JFcs0kGnohb66Rc/ +AuwmrSqdn3S/2HpyiI9OjBiYkElwn3vxHN9YKQMYB4nF8R8BqaegkBReaCdCJS0RwAc RkrPFod32Ir+Mz346WaSxGSiR6TAII4Dyy/hdz08uIUtLEQmOmbizBbdb9WSMMhZnP5l IXVXGW+yb7ACJSSHt0a5hbxV2DkwmaWr0aWPlI5dr4FZ/4GKc0PpVpIj58t5d9etVpx5 CgAw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=ECKUt25P3rRJAm43wEfY7GKrEVvZnxb1Bva2kHWnE98=; b=EHU+o8RDZ8AhxTgdFKwKJDv0EJ2Sc6Ad9wyQTAKraCub82kCC6AVtMYn4MT/1FmqPv b65bIqvbauBUDUu+26j8i6Y+bHGZf0h4diwUG/1y3J0Gz0nOgslXWjTwu/RYOH4nenAk qROlBH5FREEedh8j+KDdB6ZkTmBH+0FZkHGqDhuK8C+g7LxsUV2GrqCFAoST6wO1Qv79 eS5MCITvQ0H7oaFBj4Ci9Z7dD+gUheFjRqEw8PNiw549vBaGd8riVggWpD4iv42X5R3E SgDUHg0tzbRuxiE2/Oswrk7AJa9ibYIl/gVzIJyueVawgj54hWRkoKLGn4qcAY0CAHmU ZOmw== X-Gm-Message-State: AOAM5329Km553uJzzS3bN41YJT68d680Lp44V7M1WiYJ9CeF13WPRApP i/g5/jUnzNPP4+v4sRRyietBvT89Z95q41wBAv8zsQ== X-Google-Smtp-Source: ABdhPJzdDFHkfDJMW5s05fHeNdpulHfbS0ny8FbP06/Y+ii6+2p+Hldo8tMcVfx5b1lNjkvH04/+MIf1JMiqdqADd0w= X-Received: by 2002:a05:6512:1154:: with SMTP id m20mr322444lfg.221.1644856376137; Mon, 14 Feb 2022 08:32:56 -0800 (PST) MIME-Version: 1.0 References: <4AFF4380-4AA5-47C5-8BF0-440101A3D788@gmx.de> In-Reply-To: <4AFF4380-4AA5-47C5-8BF0-440101A3D788@gmx.de> From: =?UTF-8?Q?Bj=C3=B8rn_Ivar_Teigen?= Date: Mon, 14 Feb 2022 16:32:44 +0000 Message-ID: To: Sebastian Moeller Cc: =?UTF-8?Q?Dave_T=C3=A4ht?= , starlink@lists.bufferbloat.net Content-Type: multipart/alternative; boundary="00000000000001ae5905d7fcf8c2" Subject: Re: [Starlink] Network quality for rocket scientists X-BeenThere: starlink@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: "Starlink has bufferbloat. Bad." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Feb 2022 16:32:57 -0000 --00000000000001ae5905d7fcf8c2 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Sebastian, thank you for the feedback. "See the problem? Buffers produce jitter!" > > > I would rephrase that as "over-sized and under-managed" buffers increase > jitter unduly. Interestingly, bound jitter can be converted into static > latency by using, wait for it, (delay) buffers and a scheduler. > You are absolutely right. I was hoping someone would spot that! There is one problem with that approach though. To actually remove the jitter the scheduler can no longer be work-conserving (needs delay as you also point out), and that increases TCP ramp-up times (among other things). Can be a hard sell. I think the benefits outweigh the costs though, so I would do it your way. > > > Only nit is at the conclusion... the benefits of applying fair queuing > > to level out apparent jitter is well demonstrated at this point, for > > most kinds of traffic. Wish you'd mentioned it. > > I rather wish low latency DOCSIS (LLD) aka L4S's primary > driver/use-case would not have been mentioned before it has been properly > tested and confirmed delivering on its promises. Because most of what L4S > offers/mandates for jitter reduction is "hopes and prayers", aka end-poin= ts > are supposed to behave well... in fairness LLD at least has a poor man's = FQ > on board in the guise of "queue protection" (with by default like 32 > "buckets")... some of the cost of FQ with only few of its benefits... > Fair enough, time will show! > Regards > Sebastian > > > > > > > > > On Mon, Feb 14, 2022 at 7:07 AM Bj=C3=B8rn Ivar Teigen > wrote: > >> > >> Hi everyone, > >> > >> I was inspired by the latest Starship presentation to write a piece on > network quality in the language of rocket science. The blog can be found > here: > https://www.domos.no/news-updates/network-quality-for-rocket-scientists > >> > >> Cheers, > >> Bj=C3=B8rn Ivar Teigen > >> > >> -- > >> Bj=C3=B8rn Ivar Teigen > >> Head of Research > >> +47 47335952 | bjorn@domos.no | www.domos.no > >> WiFi Slicing by Domos > >> _______________________________________________ > >> Starlink mailing list > >> Starlink@lists.bufferbloat.net > >> https://lists.bufferbloat.net/listinfo/starlink > > > > > > > > -- > > I tried to build a better future, a few times: > > https://wayforward.archive.org/?site=3Dhttps%3A%2F%2Fwww.icei.org > > > > Dave T=C3=A4ht CEO, TekLibre, LLC > > _______________________________________________ > > Starlink mailing list > > Starlink@lists.bufferbloat.net > > https://lists.bufferbloat.net/listinfo/starlink > > --=20 Bj=C3=B8rn Ivar Teigen Head of Research +47 47335952 | bjorn@domos.no | www.domos.no WiFi Slicing by Domos --00000000000001ae5905d7fcf8c2 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Hi Sebastian, thank you for the feedback.
<= br>
"See the problem? Buffers produce jitter!"


I would rephrase that as "over-sized and under-managed" buffers i= ncrease jitter unduly. Interestingly, bound jitter can be converted into st= atic latency by using, wait for it, (delay) buffers and a scheduler.

You are absolutely right. I was hoping som= eone would spot that! There is one problem with that approach though. To ac= tually remove the jitter the scheduler can no longer be work-conserving (ne= eds delay as you also point out), and that increases TCP ramp-up times (amo= ng other things). Can be a hard sell. I think the benefits outweigh the cos= ts though, so I would do it your way.

>
> Only nit is at the conclusion... the benefits of applying fair queuing=
> to level out apparent jitter is well demonstrated at this point, for > most kinds of traffic. Wish you'd mentioned it.

=C2=A0 =C2=A0 =C2=A0 =C2=A0 I rather wish low latency DOCSIS (LLD) aka L4S&= #39;s primary driver/use-case would not have been mentioned before it has b= een properly tested and confirmed delivering on its promises. Because most = of what L4S offers/mandates for jitter reduction is "hopes and prayers= ", aka end-points are supposed to behave well... in fairness LLD at le= ast has a poor man's FQ on board in the guise of "queue protection= " (with by default like 32 "buckets")... some of the cost of= FQ with only few of its benefits...
Fair enough, time= will show!
=C2=A0
Regards
=C2=A0 =C2=A0 =C2=A0 =C2=A0 Sebastian



>
>
> On Mon, Feb 14, 2022 at 7:07 AM Bj=C3=B8rn Ivar Teigen <bjorn@domos.no> wrote:
>>
>> Hi everyone,
>>
>> I was inspired by the latest Starship presentation to write a piec= e on network quality in the language of rocket science. The blog can be fou= nd here: https://www.domos.no= /news-updates/network-quality-for-rocket-scientists
>>
>> Cheers,
>> Bj=C3=B8rn Ivar Teigen
>>
>> --
>> Bj=C3=B8rn Ivar Teigen
>> Head of Research
>> +47 47335952 | bjorn@domos.no | www.domos.no
>> WiFi Slicing by Domos
>> _______________________________________________
>> Starlink mailing list
>> Starlink@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/starl= ink
>
>
>
> --
> I tried to build a better future, a few times:
> https://wayforward.archive.org= /?site=3Dhttps%3A%2F%2Fwww.icei.org
>
> Dave T=C3=A4ht CEO, TekLibre, LLC
> _______________________________________________
> Starlink mailing list
> St= arlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink<= /a>



--
--00000000000001ae5905d7fcf8c2--