From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wr0-x230.google.com (mail-wr0-x230.google.com [IPv6:2a00:1450:400c:c0c::230]) (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 92E113B29E for ; Fri, 5 May 2017 15:29:18 -0400 (EDT) Received: by mail-wr0-x230.google.com with SMTP id z52so8995414wrc.2 for ; Fri, 05 May 2017 12:29:18 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=6jZWyebIYAGuQxDgLBel7kolYec0AuWuJr6Y+srPUH8=; b=uTbn6XKls5frsRAX4eaR7hrfQvphKcDIMbgzCfP9wf0TnrxjArGWQdIFvK1+L81nU6 1lcp9zvHj23fNEhNUT+fmuly6Dmmfbzrli7KoYC4JOMu966el5a8Npr0oVr4FF1qFdxR FvW09qkerlNEyx/Rh0wsUl24S9K8+J1fthuLaucdIuSx10bL5UVxT2Yr2ymAbVRmZKA3 FexvqrXsiDso/viBKt0EIovcLDDPeLldt/rIx37ctdc2enrTYqBTLEeSRtpX8oo5QeaT f2u/eUY5VyJr+/gq6GNg20ybKiZk5Csb18orhxGB05+DDD0q8NGpdOeKodxv07XEvPeW lpJQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=6jZWyebIYAGuQxDgLBel7kolYec0AuWuJr6Y+srPUH8=; b=cBp/iU6G2V1obiw72CEbaQy1imDaw228DhELMxvIs/M177otQDsQC36ufgM0990AvZ oHY7YmlARbhlMTzZ9zMRE3rPXWwYrCVCXi2vhVT7NnKZ+3RQ4UN9ZV3VyF4H0+nHO3fr Ix6yLS1UsccYhpwTWGLsvH93PCEAd5fr1QjvilNnqrI+T/uESIIWRKdCknjYoEQftdv+ gR7+u0xuwd1HD5n43h6kjBWh4lj/DibVjeP+LTbPjcYQInDr2uLboboxbFLC3SnTtR1H lD3u6jC1oefEvBTxnztsys8bMQ8P2Ns5gf/WTqCyKnXPgqWldprFqoka4gdKqA/6SV3q SNAw== X-Gm-Message-State: AN3rC/5jiRi85V/yytP9BlzaVTVGFB0vYX6hz2KtOC6IYgowRB3fQZka Au7ddkz1Gte8wQrnKdIY+wPUg2ZZtA== X-Received: by 10.223.171.79 with SMTP id r15mr36297704wrc.149.1494012557501; Fri, 05 May 2017 12:29:17 -0700 (PDT) MIME-Version: 1.0 References: <1493989225450.12354@telenor.com> <1493992526496.9641@telenor.com> <87wp9vjtex.fsf@alrua-kau> <0F8A4A7F-2CD1-4DEF-99A9-1E14ECDA5D2D@telenor.com> <661B553B-512D-4210-9CDF-0ADD1B21CE0C@toke.dk> <1494004366992.6614@telenor.com> In-Reply-To: <1494004366992.6614@telenor.com> From: Neil Shepperd Date: Fri, 05 May 2017 19:29:06 +0000 Message-ID: To: erik.taraldsen@telenor.com, toke@toke.dk Cc: cake@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=94eb2c1cb818c7ab24054ecbe79f Subject: Re: [Cake] ER-X now running cake, thanks for the help. :) 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: Fri, 05 May 2017 19:29:18 -0000 --94eb2c1cb818c7ab24054ecbe79f Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Shouldn't unresponsive UDP flows be punished for sending too many packets? In that case if there's any problem at all it is that the packets get through with 20 seconds delay instead of being dropped entirely. Although it shouldn't really matter either way as long as other flows are not unfairly impacted. On Fri, 5 May 2017 at 13:12 wrote: > > Fra: Toke H=C3=B8iland-J=C3=B8rgensen > > > > Right, so you're measuring the actual latency experienced by the UDP > packets in the > > unresponsive flow? Yeah, you'll get quite a bit of latency in that case= . > But not sure that's a case > > we should be optimising for... > > That was just the worst case scenario, not really meant to emulate user > traffic. And I'm not suggesting that the schedulers should be optimized > for that. But we can agree that 20+ seconds of delay is a bit much no > matter how it is measured? :) > > I also use nuttcp to model other kinds of traffic, such as VIOP, Voice > over WiFi, dumb udp vod that Telenor uses. Then I subject the router to > various kinds of line saturation and stress. Cloud upload, torrent, viru= s > emulation (zmap not really designed for that, but man, if you want to > inflict pain on a router... zmap is your daddy). And we try to make sure > that long living tcp/vpn sessions are not torn down due to need for more > sessions in nat for yet another torrent session. Different QoS/scheduler= s > behave different and we try and use the best possible solution for a give= n > HW/access tech. What works very well for balancing cloud upload and > interactive traffic like ssh or http may fail miserably on torrent. And > the other way around. On our legacy stuff we can't implement fq_codel, > cake or sfq even. > > And as you know I also incorporate flent. > > > -Erik > _______________________________________________ > Cake mailing list > Cake@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/cake > --94eb2c1cb818c7ab24054ecbe79f Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Shouldn't unresponsive UDP flows be punished for sendi= ng too many packets? In that case if there's any problem at all it is t= hat the packets get through with 20 seconds delay instead of being dropped = entirely. Although it shouldn't really matter either way as long as oth= er flows are not unfairly impacted.

On Fri, 5 May 2017 at 13:12 <erik.taraldsen@telenor.com> wrote:
> Fra: Toke H=C3=B8iland-J=C3=B8rgensen <toke@toke.dk>
>
> Right, so you're measuring the actual latency experienced by the U= DP packets in the
> unresponsive flow? Yeah, you'll get quite a bit of latency in that= case. But not sure that's a case
> we should be optimising for...

That was just the worst case scenario, not really meant to emulate user tra= ffic.=C2=A0 And I'm not suggesting that the schedulers should be optimi= zed for that. But we can agree that 20+ seconds of delay is a bit much no m= atter how it is measured? :)

I also use nuttcp to model other kinds of traffic, such as VIOP, Voice over= WiFi, dumb udp vod that Telenor uses.=C2=A0 Then I subject the router to v= arious kinds of line saturation and stress.=C2=A0 Cloud upload, torrent, vi= rus emulation (zmap not really designed for that, but man, if you want to i= nflict pain on a router... zmap is your daddy).=C2=A0 And we try to make su= re that long living tcp/vpn sessions are not torn down due to need for more= sessions in nat for yet another torrent session.=C2=A0 Different QoS/sched= ulers behave different and we try and use the best possible solution for a = given HW/access tech.=C2=A0 What works very well for balancing cloud upload= and interactive traffic like ssh or http may fail miserably on torrent.=C2= =A0 And the other way around.=C2=A0 On our legacy stuff we can't implem= ent fq_codel, cake or sfq even.

And as you know I also incorporate flent.


-Erik
_______________________________________________
Cake mailing list
Cake@lists.= bufferbloat.net
https://lists.bufferbloat.net/listinfo/cake
--94eb2c1cb818c7ab24054ecbe79f--