From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qt1-x829.google.com (mail-qt1-x829.google.com [IPv6:2607:f8b0:4864:20::829]) (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 60FA73BA8E for ; Wed, 28 Nov 2018 05:40:35 -0500 (EST) Received: by mail-qt1-x829.google.com with SMTP id r14so25265922qtp.1 for ; Wed, 28 Nov 2018 02:40:35 -0800 (PST) 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:content-transfer-encoding; bh=N84xdayGOd52lXvGr5Z3HWFje6eOVm3eLlwi2/CwGpc=; b=I/ir4kg7gZgFZ5rQ/X9Ub0Fs+67Q3ImWQHn/pRlT510tWJjnwMxCVuNrkzFfGG54+P kSdLJp/W0FGKzx+aRSctWPppjSo3iBu2deBgDVu/mF5ewEUaQL8WlTTyAuJ2l6XCc95l J5pmxtIiK3l1Lc0IT0e3oROx5qKhQUUQ5XMDxjlSJLm0rTlkcYXwrto9jDYqgofT+mAT JV8MPAP8Cnb/fRNZ0wBF5BUj+i7/8xMAvV9SDOPgAd4Ie0sLBWLIyuL8Q/qB0nGxZEZl 8pGvPUpO+rXqCxfGsjxNbRsdnVBEL4QmMKWmiXkHYX0H5K0WGpxw1E8UWrzCgDqEh9HO Kvpw== 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:content-transfer-encoding; bh=N84xdayGOd52lXvGr5Z3HWFje6eOVm3eLlwi2/CwGpc=; b=BRFfDgU24FtC3rI0RYwH0VDf2ZKHJhPF36eD6ATL+D8nAEjkJXGYudxQ73EWZeSLf6 Qh3dmq1sHVpWGtlU7k/kYg4w9K9qtW2NLmJMO13cy2Ms/LrdRpLwkCT6SrBeAGfilQjO x4Yox9nnY3J3Or7O0JQBbzLbCuackoZjGDB/LYtfZHsO7/xPSwWfTOufCvura3JOvcdc SsCnoAaR/qqcEqkGXU31YwTNkQJeedMjMz9zrOU4rVLamk1EKNqv3yboDPHyhmWpfmMH eM0ZOiZudqm9n2+OnpErLn5MMbHGfUpzxOEA4GuH+Ns1LsJZBviVPi66MwUJH8oEaUO8 xggA== X-Gm-Message-State: AA+aEWbu22oSdWUx0YZOTaeZZ37P8erksB/nDy6MVGZfRnduKvQ14iFK S+mdvRSl9OFsz2t9WPD02Bp1pxUzd8ZxWQOsKXA= X-Google-Smtp-Source: AFSGD/V7MHJ5+NibHiAHmhGSes0ce/tNyUv4eWGyBLIICVqf0DSsBBuYck6A92q2lNVDw0wKHQ0wCBYAPwPe5Jpp8yo= X-Received: by 2002:a0c:afd1:: with SMTP id t17mr34397250qvc.93.1543401634730; Wed, 28 Nov 2018 02:40:34 -0800 (PST) MIME-Version: 1.0 References: <65EAC6C1-4688-46B6-A575-A6C7F2C066C5@heistp.net> In-Reply-To: From: Dave Taht Date: Wed, 28 Nov 2018 02:40:23 -0800 Message-ID: To: Luca MUSCARIELLO Cc: Jonathan Morton , bloat Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Subject: Re: [Bloat] when does the CoDel part of fq_codel help in the real world? 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: Wed, 28 Nov 2018 10:40:35 -0000 On Wed, Nov 28, 2018 at 1:56 AM Luca Muscariello wrote: > > Dave, > > The single BDP inflight is a rule of thumb that does not account for fluc= tuations of the RTT. > And I am not talking about random fluctuations and noise. I am talking ab= out fluctuations > from a control theoretic point of view to stabilise the system, e.g. the = trajectory of the system variable that > gets to the optimal point no matter the initial conditions (Lyapunov). I have been trying all day to summon the gumption to make this argument: IF you have a good idea of the actual RTT... it is also nearly certain that there will be *at least* one other flow you will be competing with... therefore the fluctuations from every point of view are dominated by the interaction between these flows and the goal is, in general, is not to take up a full BDP for your single flow. And BBR aims for some tiny percentage less than what it thinks it can get, when, well, everybody's seen it battle it out with itself and with cubic. I hand it FQ at the bottleneck link and it works well. single flows exist only in the minds of theorists and labs. There's a relevant passage worth citing in the kleinrock paper, I thought (did he write two recently?) that talked about this problem... I *swear* when I first read it it had a deeper discussion of the second sentence below and had two paragraphs that went into the issues with multiple flows: "ch earlier and led to the Flow Deviation algorithm [28]. 17 The reason that the early work of 40 years ago took so long to make its current impact is because in [31] it was shown that the mechanism presented in [2] and [3] could not be implemented in a decentralized algorithm. This delayed the application of Power until the recent work by the Google team in [1] demonstrated that the key elements of response time and bandwidth could indeed be estimated using a distributed control loop sliding window spanning approximately 10 round-trip times." but I can't find it today. > The ACM queue paper talking about Codel makes a fairly intuitive and acce= ssible explanation of that. I haven't re-read the lola paper. I just wanted to make the assertion above. And then duck. :) Also, when I last looked at BBR, it made a false assumption that 200ms was "long enough" to probe the actual RTT, when my comcast links and others are measured at 680ms+ of buffering. And I always liked the stanford work, here, which tried to assert that a link with n flows requires no more than B =3D (RTT =C3=97C)/ =E2=88=9A n. http://yuba.stanford.edu/techreports/TR04-HPNG-060800.pdf night! > There is a less accessible literature talking about that, which dates bac= k to some time ago > that may be useful to re-read again > > Damon Wischik and Nick McKeown. 2005. > Part I: buffer sizes for core routers. > SIGCOMM Comput. Commun. Rev. 35, 3 (July 2005), 75-78. DOI=3Dhttp://dx.do= i.org/10.1145/1070873.1070884 > http://klamath.stanford.edu/~nickm/papers/BufferSizing.pdf.pdf > > and > > Gaurav Raina, Don Towsley, and Damon Wischik. 2005. > Part II: control theory for buffer sizing. > SIGCOMM Comput. Commun. Rev. 35, 3 (July 2005), 79-82. > DOI=3Dhttp://dx.doi.org/10.1145/1070873.1070885 > http://www.statslab.cam.ac.uk/~gr224/PAPERS/Control_Theory_Buffers.pdf > > One of the thing that Frank Kelly has brought to the literature is about = optimal control. > From a pure optimization point of view we know since Robert Gallagher (an= d Bertsekas 1981) that > the optimal sending rate is a function of the shadow price at the bottlen= eck. > This shadow price is nothing more than the Lagrange multiplier of the cap= acity constraint > at the bottleneck. Some protocols such as XCP or RCP propose to carry som= ething > very close to a shadow price in the ECN but that's not that simple. > And currently we have a 0/1 "shadow price" which way insufficient. > > Optimal control as developed by Frank Kelly since 1998 tells you that you= have > a stability region that is needed to get to the optimum. > > Wischik work, IMO, helps quite a lot to understand tradeoffs while design= ing AQM > and CC. I feel like the people who wrote the codel ACM Queue paper are ve= ry much aware of this literature, > because Codel design principles seem to take into account that. > And the BBR paper too. > > > On Tue, Nov 27, 2018 at 9:58 PM Dave Taht wrote: >> >> OK, wow, this conversation got long. and I'm still 20 messages behind. >> >> Two points, and I'm going to go back to work, and maybe I'll try to >> summarize a table >> of the competing viewpoints, as there's far more than BDP of >> discussion here, and what >> we need is sqrt(bdp) to deal with all the different conversational flows= . :) >> >> On Tue, Nov 27, 2018 at 1:24 AM Luca Muscariello >> wrote: >> > >> > I think that this is a very good comment to the discussion at the defe= nse about the comparison between >> > SFQ with longest queue drop and FQ_Codel. >> > >> > A congestion controlled protocol such as TCP or others, including QUIC= , LEDBAT and so on >> > need at least the BDP in the transmission queue to get full link effic= iency, i.e. the queue never empties out. >> >> no, I think it needs a BDP in flight. >> >> I think some of the confusion here is that your TCP stack needs to >> keep around a BDP in order to deal with >> retransmits, but that lives in another set of buffers entirely. >> >> > This gives rule of thumbs to size buffers which is also very practical= and thanks to flow isolation becomes very accurate. >> > >> > Which is: >> > >> > 1) find a way to keep the number of backlogged flows at a reasonable v= alue. >> > This largely depends on the minimum fair rate an application may need = in the long term. >> > We discussed a little bit of available mechanisms to achieve that in t= he literature. >> > >> > 2) fix the largest RTT you want to serve at full utilization and size = the buffer using BDP * N_backlogged. >> > Or the other way round: check how much memory you can use >> > in the router/line card/device and for a fixed N, compute the largest = RTT you can serve at full utilization. >> >> My own take on the whole BDP argument is that *so long as the flows in >> that BDP are thoroughly mixed* you win. >> >> > >> > 3) there is still some memory to dimension for sparse flows in additio= n to that, but this is not based on BDP. >> > It is just enough to compute the total utilization of sparse flows and= use the same simple model Toke has used >> > to compute the (de)prioritization probability. >> > >> > This procedure would allow to size FQ_codel but also SFQ. >> > It would be interesting to compare the two under this buffer sizing. >> > It would also be interesting to compare another mechanism that we have= mentioned during the defense >> > which is AFD + a sparse flow queue. Which is, BTW, already available i= n Cisco nexus switches for data centres. >> > >> > I think that the the codel part would still provide the ECN feature, t= hat all the others cannot have. >> > However the others, the last one especially can be implemented in sili= con with reasonable cost. >> > >> > >> > >> > >> > >> > On Mon 26 Nov 2018 at 22:30, Jonathan Morton w= rote: >> >> >> >> > On 26 Nov, 2018, at 9:08 pm, Pete Heist wrote: >> >> > >> >> > So I just thought to continue the discussion- when does the CoDel p= art of fq_codel actually help in the real world? >> >> >> >> Fundamentally, without Codel the only limits on the congestion window= would be when the sender or receiver hit configured or calculated rwnd and= cwnd limits (the rwnd is visible on the wire and usually chosen to be larg= e enough to be a non-factor), or when the queue overflows. Large windows r= equire buffer memory in both sender and receiver, increasing costs on the s= ender in particular (who typically has many flows to manage per machine). >> >> >> >> Queue overflow tends to result in burst loss and head-of-line blockin= g in the receiver, which is visible to the user as a pause and subsequent j= ump in the progress of their download, accompanied by a major fluctuation i= n the estimated time to completion. The lost packets also consume capacity= upstream of the bottleneck which does not contribute to application throug= hput. These effects are independent of whether overflow dropping occurs at= the head or tail of the bottleneck queue, though recovery occurs more quic= kly (and fewer packets might be lost) if dropping occurs from the head of t= he queue. >> >> >> >> From a pure throughput-efficiency standpoint, Codel allows using ECN = for congestion signalling instead of packet loss, potentially eliminating p= acket loss and associated lead-of-line blocking entirely. Even without ECN= , the actual cwnd is kept near the minimum necessary to satisfy the BDP of = the path, reducing memory requirements and significantly shortening the rec= overy time of each loss cycle, to the point where the end-user may not noti= ce that delivery is not perfectly smooth, and implementing accurate complet= ion time estimators is considerably simplified. >> >> >> >> An important use-case is where two sequential bottlenecks exist on th= e path, the upstream one being only slightly higher capacity but lacking an= y queue management at all. This is presently common in cases where home CP= E implements inbound shaping on a generic ISP last-mile link. In that case= , without Codel running on the second bottleneck, traffic would collect in = the first bottleneck's queue as well, greatly reducing the beneficial effec= ts of FQ implemented on the second bottleneck. In this topology, the overa= ll effect is inter-flow as well as intra-flow. >> >> >> >> The combination of Codel with FQ is done in such a way that a separat= e instance of Codel is implemented for each flow. This means that congesti= on signals are only sent to flows that require them, and non-saturating flo= ws are unmolested. This makes the combination synergistic, where each comp= onent offers an improvement to the behaviour of the other. >> >> >> >> - Jonathan Morton >> >> >> >> _______________________________________________ >> >> Bloat mailing list >> >> Bloat@lists.bufferbloat.net >> >> https://lists.bufferbloat.net/listinfo/bloat >> > >> > _______________________________________________ >> > Bloat mailing list >> > Bloat@lists.bufferbloat.net >> > https://lists.bufferbloat.net/listinfo/bloat >> >> >> >> -- >> >> Dave T=C3=A4ht >> CTO, TekLibre, LLC >> http://www.teklibre.com >> Tel: 1-831-205-9740 -- Dave T=C3=A4ht CTO, TekLibre, LLC http://www.teklibre.com Tel: 1-831-205-9740