One document matched: draft-ietf-tsvwg-byte-pkt-congest-02.xml


<?xml version="1.0" encoding="US-ASCII"?>
<!DOCTYPE rfc SYSTEM "rfc2629.dtd">
<rfc category="info" docName="draft-ietf-tsvwg-byte-pkt-congest-02"
     ipr="trust200902" updates="2309">
  <?xml-stylesheet type='text/xsl' href='http://xml.resource.org/authoring/rfc2629.xslt' ?>

  <!-- Alterations to I-D/RFC boilerplate -->

  <?rfc private="" ?>

  <!-- Default private="" Produce an internal memo 2.5pp shorter than an I-D or RFC -->

  <?rfc topblock="yes" ?>

  <!-- Default topblock="yes" put the famous header block on the first page -->

  <?rfc footer="" ?>

  <!-- Default footer="Expires <date>" override the center footer string -->

  <?rfc header="" ?>

  <!-- Default header="Internet-Draft" override the leftmost header string -->

  <?rfc authorship="yes" ?>

  <!-- Default authorship="yes" Render authors' addresses section -->

  <?rfc rfcprocack="yes" ?>

  <!-- Default rfcprocack="no" add a short sentence acknowledging xml2rfc -->

  <?rfc strict="no" ?>

  <!-- Default strict="no" Don't check I-D nits -->

  <?rfc rfcedstyle="yes" ?>

  <!-- Default rfcedstyle="yes" attempt to closely follow finer details from the latest observable RFC-Editor style -->

  <?rfc colonspace="no" ?>

  <!-- Default colonspace="no" put two spaces instead of one after each colon (``:'') in txt or nroff files -->

  <?rfc notedraftinprogress="yes" ?>

  <!-- Default notedraftinprogress="yes" generates "(work in progress)", as appropriate -->

  <?rfc refparent="References" ?>

  <!-- Default refparent="References" title of the top-level section containing all references -->

  <!-- IETF process -->

  <?rfc iprnotified="no" ?>

  <!-- Default iprnotified="no" I haven't disclosed existence of IPR to IETF -->

  <!-- ToC format -->

  <?rfc toc="yes" ?>

  <!-- Default toc="no" No Table of Contents -->

  <?rfc tocappendix="yes" ?>

  <!-- Default tocappendix="yes" control whether the word `Appendix' appears in the table-of-content -->

  <?rfc tocdepth="3" ?>

  <!-- Default tocdepth="3" if toc is "yes", then this determines the depth of the table-of-contents -->

  <?rfc tocindent="yes" ?>

  <!-- Default tocindent="yes" if toc is "yes", indent subsections in the table-of-contents -->

  <?rfc tocnarrow="yes" ?>

  <!-- Default tocnarrow="yes" affects horizontal spacing in the table-of-content -->

  <?rfc tocompact="yes" ?>

  <!-- Default tocompact="yes" if toc is "yes", then setting this to "no" will make it a little less compact -->

  <!-- Cross referencing, footnotes, comments -->

  <?rfc symrefs="yes" ?>

  <!-- Default symrefs="no" Don't use anchors, but use numbers for refs -->

  <?rfc sortrefs="yes"?>

  <!-- Default sortrefs="no" Don't sort references into order -->

  <?rfc comments="no" ?>

  <!-- Default comments="yes" Don't render comments -->

  <?rfc inline="no" ?>

  <!-- Default inline="no" if comments is "yes", then render comments inline; otherwise render them in an `Editorial Comments' section -->

  <?rfc editing="no" ?>

  <!-- Default editing="no" Don't insert editing marks for ease of discussing draft versions -->

  <!-- Pagination control -->

  <?rfc compact="yes"?>

  <!-- Default compact="no" Start sections on new pages -->

  <?rfc subcompact="no"?>

  <!-- Default subcompact="(as compact setting)" yes/no is not quite as compact as yes/yes -->

  <?rfc autobreaks="yes" ?>

  <!-- Default autobreaks="yes" avoid widows and orphans (not perfect) -->

  <!-- HTML formatting control -->

  <?rfc emoticonic="yes" ?>

  <!-- Default emoticonic="no" Doesn't prettify HTML format -->

  <?rfc background="" ?>

  <!-- Default background="" when producing a html file, use this image -->

  <?rfc useobject="no" ?>

  <!-- Default useobject="no" use <object> not <src> when outputting HTML -->

  <?rfc linkmailto="yes" ?>

  <!-- Default linkmailto="yes" generate mailto: URL, as appropriate -->

  <?rfc docmapping="no" ?>

  <!-- Default docmapping="no" use hierarchical tags (e.g., <h1>, <h2>, etc.) for (sub)section titles -->

  <?rfc slides="no" ?>

  <!-- Default slides="no" when producing a html file, produce multiple files for a slide show -->

  <front>
    <title abbrev="Byte and Packet Congestion Notification">Byte and Packet
    Congestion Notification</title>

    <author fullname="Bob Briscoe" initials="B." surname="Briscoe">
      <organization>BT</organization>

      <address>
        <postal>
          <street>B54/77, Adastral Park</street>

          <street>Martlesham Heath</street>

          <city>Ipswich</city>

          <code>IP5 3RE</code>

          <country>UK</country>
        </postal>

        <phone>+44 1473 645196</phone>

        <email>bob.briscoe@bt.com</email>

        <uri>http://bobbriscoe.net/</uri>
      </address>
    </author>

    <author fullname="Jukka Manner" initials="J." surname="Manner">
      <organization abbrev="Aalto University">Aalto University</organization>

      <address>
        <postal>
          <street>Department of Communications and Networking
          (Comnet)</street>

          <street>P.O. Box 13000</street>

          <code>FIN-00076 Aalto</code>

          <country>Finland</country>
        </postal>

        <phone>+358 9 470 22481</phone>

        <email>jukka.manner@tkk.fi</email>

        <uri>http://www.netlab.tkk.fi/~jmanner/</uri>
      </address>
    </author>

    <date day="12" month="July" year="2010" />

    <area>Transport</area>

    <workgroup>Transport Area Working Group</workgroup>

    <keyword>Active queue management (AQM)</keyword>

    <keyword>Availability</keyword>

    <keyword>Denial of Service</keyword>

    <keyword>Quality of Service (QoS)</keyword>

    <keyword>Congestion Control</keyword>

    <keyword>Fairness</keyword>

    <keyword>Incentives</keyword>

    <keyword>Protocol</keyword>

    <keyword>Architecture layering</keyword>

    <abstract>
      <t>This memo concerns dropping or marking packets using active queue
      management (AQM) such as random early detection (RED) or pre-congestion
      notification (PCN). We give two strong recommendations: (1) packet size
      should not be taken into account when transports read congestion
      indications, not when network equipment writes them, and (2) byte-mode
      packet drop variant of AQM algorithms, such as RED, should not be used
      to drop fewer small packets.</t>
    </abstract>
  </front>

  <middle>
    <section anchor="pktb_Introduction" title="Introduction">
      <t>When notifying congestion, the problem of how (and whether) to take
      packet sizes into account has exercised the minds of researchers and
      practitioners for as long as active queue management (AQM) has been
      discussed. Indeed, one reason AQM was originally introduced was to
      reduce the lock-out effects that small packets can have on large packets
      in drop-tail queues. This memo aims to state the principles we should be
      using and to come to conclusions on what these principles will mean for
      future protocol design, taking into account the deployments we have
      already.</t>

      <t>The byte vs. packet dilemma arises at three stages in the congestion
      notification process: <list style="hanging">
          <t hangText="Measuring congestion:">When the congested resource
          decides locally to measure how congested it is. (Should the queue
          measure its length in bytes or packets?);</t>

          <t
          hangText="Coding congestion notification into the wire protocol:">When
          the congested resource decides whether to notify the level of
          congestion on each particular packet. (When a queue considers
          whether to notify congestion by dropping or marking a particular
          packet, should its decision depend on the byte-size of the
          particular packet being dropped or marked?);</t>

          <t
          hangText="Decoding congestion notification from the wire protocol:">When
          the transport interprets the notification in order to decide how
          much to respond to congestion. (Should the transport take into
          account the byte-size of each missing or marked packet?).</t>
        </list></t>

      <t>Consensus has emerged over the years concerning the first stage:
      whether queues are measured in bytes or packets, termed byte-mode queue
      measurement or packet-mode queue measurement. This memo records this
      consensus in the RFC Series. In summary the choice solely depends on
      whether the resource is congested by bytes or packets.</t>

      <t>The controversy is mainly around the last two stages to do with
      encoding congestion notification into packets: whether to allow for the
      size of the specific packet notifying congestion i) when the network
      encodes or ii) when the transport decodes the congestion
      notification.</t>

      <t>Currently, the RFC series is silent on this matter other than a paper
      trail of advice referenced from <xref target="RFC2309"></xref>, which
      conditionally recommends byte-mode (packet-size dependent) drop <xref
      target="pktByteEmail"></xref>. The primary purpose of this memo is to
      build a definitive consensus against such deliberate preferential
      treatment for small packets in AQM algorithms and to record this advice
      within the RFC series. Fortunately all the implementers who responded to
      our survey (<xref target="pktb_Coding_Status_Summary"></xref>) have not
      followed the earlier advice, so the consensus this memo argues for seems
      to already exist in implementations. </t>

      <t>The primary conclusion of this memo is that packet size should be
      taken into account when transports read congestion indications, not when
      network equipment writes them. Reducing drop of small packets has some
      tempting advantages: i) it drops less control packets, which tend to be
      small and ii) it makes TCP's bit-rate less dependent on packet size.
      However, there are ways of addressing these issues at the transport
      layer, rather than reverse engineering network forwarding to fix
      specific transport problems.</t>

      <t>The second conclusion is that network layer algorithms like the
      byte-mode packet drop variant of RED should not be used to drop fewer
      small packets, because that creates a perverse incentive for transports
      to use tiny segments, consequently also opening up a DoS
      vulnerability.</t>

      <t>This memo is initially concerned with how we should correctly scale
      congestion control functions with packet size for the long term. But it
      also recognises that expediency may be necessary to deal with existing
      widely deployed protocols that don't live up to the long term goal. It
      turns out that the 'correct' variant of RED to deploy seems to be the
      one everyone has deployed, and no-one who responded to our survey has
      implemented the other variant. However, at the transport layer, TCP
      congestion control is a widely deployed protocol that we argue doesn't
      scale correctly with packet size. To date this hasn't been a significant
      problem because most TCPs have been used with similar packet sizes. But,
      as we design new congestion controls, we should build in scaling with
      packet size rather than assuming we should follow TCP's example.</t>

      <!--
Then the body of the memo starts from first
      principles, defining congestion notification in <xref
      target="pktb_Congestion_Definition" /> then determining the correct way
      to measure congestion (<xref target="pktb_Measure" />) and to design an
      idealised congestion notification protocol (<xref
      target="pktb_Ideal_Coding" />). 
-->

      <t>This memo continues as follows. Terminology and scoping are discussed
      next, and the reasons to make the recommendations presented in this memo
      now are given in <xref target="pktb_now"></xref>. Motivating arguments
      for our advice are given in <xref target="pktb_Motivation"></xref>. We
      then survey the advice given previously in the RFC series, the research
      literature and the deployed legacy (<xref target="pktb_SotA"></xref>)
      before listing outstanding issues (<xref target="pktb_Issues"></xref>)
      that will need resolution both to inform future protocols designs and to
      handle legacy. We then give concrete recommendations for the way forward
      in (<xref target="pktb_Conclusions"></xref>). We finally give security
      considerations in <xref target="pktb_Security_Considerations"></xref>.
      The interested reader can also find further discussions about the theme
      of byte vs. packet in the appendices.</t>

      <t>This memo intentionally includes a non-negligible amount of material
      on the subject. A busy reader can jump right into <xref
      target="pktb_Conclusions"></xref> to read a summary of the
      recommendations for the Internet community.</t>

      <section anchor="pktb_term" title="Terminology and Scoping">
        <t>The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
        "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
        document are to be interpreted as described in <xref
        target="RFC2119"></xref>.</t>

        <!-- Old section 3 below ================================================================ 


    <section anchor="pktb_Congestion_Definition"
             title="Working Definition of Congestion Notification">
-->

        <t><list style="hanging">
            <t hangText="Congestion Notification:">Rather than aim to achieve
            what many have tried and failed, this memo will not try to define
            congestion. It will give a working definition of what congestion
            notification should be taken to mean for this document. Congestion
            notification is a changing signal that aims to communicate the
            ratio E/L. E is the instantaneous excess load offered to a
            resource that it is either incapable of serving or unwilling to
            serve. L is the instantaneous offered load.<vspace
            blankLines="1" />The phrase `unwilling to serve' is added, because
            AQM systems (e.g. RED, PCN <xref target="RFC5670" />) set a
            virtual limit smaller than the actual limit to the resource, then
            notify when this virtual limit is exceeded in order to avoid
            congestion of the actual capacity.<vspace blankLines="1" />Note
            that the denominator is offered load, not capacity. Therefore
            congestion notification is a real number bounded by the range
            [0,1]. This ties in with the most well-understood measure of
            congestion notification: drop fraction (often loosely called loss
            rate). It also means that congestion has a natural interpretation
            as a probability; the probability of offered traffic not being
            served (or being marked as at risk of not being served). <xref
            target="pktb_CN_Definition"></xref> describes a further incidental
            benefit that arises from using load as the denominator of
            congestion notification.</t>

            <t hangText="Explicit and Implicit Notification:">The byte vs.
            packet dilemma concerns congestion notification irrespective of
            whether it is signalled implicitly by drop or using explicit
            congestion notification (ECN <xref target="RFC3168"></xref> or PCN
            <xref target="RFC5670"></xref>). Throughout this document, unless
            clear from the context, the term marking will be used to mean
            notifying congestion explicitly, while congestion notification
            will be used to mean notifying congestion either implicitly by
            drop or explicitly by marking.</t>

            <t hangText="Bit-congestible vs. Packet-congestible:">If the load
            on a resource depends on the rate at which packets arrive, it is
            called packet-congestible. If the load depends on the rate at
            which bits arrive it is called bit-congestible.<vspace
            blankLines="1" />Examples of packet-congestible resources are
            route look-up engines and firewalls, because load depends on how
            many packet headers they have to process. Examples of
            bit-congestible resources are transmission links, radio power and
            most buffer memory, because the load depends on how many bits they
            have to transmit or store. Some machine architectures use fixed
            size packet buffers, so buffer memory in these cases is
            packet-congestible (see <xref
            target="pktb_Fixed_Buffers"></xref>).<vspace
            blankLines="1" />Currently a design goal of network processing
            equipment such as routers and firewalls is to keep packet
            processing uncongested even under worst case bit rates with
            minimum packet sizes. Therefore, packet-congestion is currently
            rare, but there is no guarantee that it will not become common
            with future technology trends.<vspace blankLines="1" />Note that
            information is generally processed or transmitted with a minimum
            granularity greater than a bit (e.g. octets). The appropriate
            granularity for the resource in question should be used, but for
            the sake of brevity we will talk in terms of bytes in this
            memo.</t>

            <t hangText="Coarser granularity:">Resources may be congestible at
            higher levels of granularity than packets, for instance stateful
            firewalls are flow-congestible and call-servers are
            session-congestible. This memo focuses on congestion of
            connectionless resources, but the same principles may be
            applicable for congestion notification protocols controlling
            per-flow and per-session processing or state.</t>

            <t hangText="RED Terminology:">In RED, whether to use packets or
            bytes when measuring queues is respectively called packet-mode or
            byte-mode queue measurement. And if the probability of dropping a
            packet depends on its byte-size it is called byte-mode drop,
            whereas if the drop probability is independent of a packet's
            byte-size it is called packet-mode drop.</t>
          </list></t>
      </section>

      <section anchor="pktb_now" title="Why now?">
        <t>Now is a good time to discuss whether fairness between different
        sized packets would best be implemented in the network layer, or at
        the transport, for a number of reasons: <list style="numbers">
            <t>The packet vs. byte issue requires speedy resolution because
            the IETF pre-congestion notification (PCN) working group is
            standardising the external behaviour of a PCN congestion
            notification (AQM) algorithm <xref target="RFC5670"></xref>;</t>

            <t><xref target="RFC2309"></xref> says RED may either take account
            of packet size or not when dropping, but gives no recommendation
            between the two, referring instead to advice on the performance
            implications in an email <xref target="pktByteEmail"></xref>,
            which recommends byte-mode drop. Further, just before RFC2309 was
            issued, an addendum was added to the archived email that revisited
            the issue of packet vs. byte-mode drop in its last paragraph,
            making the recommendation less clear-cut;</t>

            <t>Without the present memo, the only advice in the RFC series on
            packet size bias in AQM algorithms would be a reference to an
            archived email in <xref target="RFC2309"></xref> (including an
            addendum at the end of the email to correct the original).</t>

            <t>The IRTF Internet Congestion Control Research Group (ICCRG)
            recently took on the challenge of building consensus on what
            common congestion control support should be required from network
            forwarding functions in future <xref
            target="I-D.irtf-iccrg-welzl" />.
            The wider Internet community needs to discuss whether the
            complexity of adjusting for packet size should be in the network
            or in transports;</t>

            <t>Given there are many good reasons why larger path max
            transmission units (PMTUs) would help solve a number of scaling
            issues, we don't want to create any bias against large packets
            that is greater than their true cost;</t>

            <t>The IETF has started to consider the question of fairness
            between flows that use different packet sizes (e.g. in the
            small-packet variant of TCP-friendly rate control, TFRC-SP <xref
            target="RFC4828"></xref>). Given transports with different packet
            sizes, if we don't decide whether the network or the transport
            should allow for packet size, it will be hard if not impossible to
            design any transport protocol so that its bit-rate relative to
            other transports meets design guidelines <xref
            target="RFC5033"></xref> (Note however that, if the concern were
            fairness between users, rather than between flows <xref
            target="Rate_fair_Dis"></xref>, relative rates between flows would
            have to come under run-time control rather than being embedded in
            protocol designs).</t>
          </list></t>
      </section>
    </section>

    <!-- ================================================================ -->

    <section anchor="pktb_Motivation" title="Motivating Arguments">
      <section anchor="pktb_Scaling"
               title="Scaling Congestion Control with Packet Size">
        <t>There are two ways of interpreting a dropped or marked packet. It
        can either be considered as a single loss event or as loss/marking of
        the bytes in the packet. Here we try to design a test to see which
        approach scales with packet size.</t>

        <t>Given bit-congestible is the more common case (see <xref
        target="pktb_term"></xref>), consider a bit-congestible link shared by
        many flows, so that each busy period tends to cause packets to be lost
        from different flows. The test compares two identical scenarios with
        the same applications, the same numbers of sources and the same load.
        But the sources break the load into large packets in one scenario and
        small packets in the other. Of course, because the load is the same,
        there will be proportionately more packets in the small packet
        case.</t>

        <t>The test of whether a congestion control scales with packet size is
        that it should respond in the same way to the same congestion
        excursion, irrespective of the size of the packets that the bytes
        causing congestion happen to be broken down into.</t>

        <t>A bit-congestible queue suffering a congestion excursion has to
        drop or mark the same excess bytes whether they are in a few large
        packets or many small packets. So for the same congestion excursion,
        the same amount of bytes have to be shed to get the load back to its
        operating point. But, of course, for smaller packets more packets will
        have to be discarded to shed the same bytes.</t>

        <t>If all the transports interpret each drop/mark as a single loss
        event irrespective of the size of the packet dropped, those with
        smaller packets will respond more to the same congestion excursion,
        failing our test. On the other hand, if they respond proportionately
        less when smaller packets are dropped/marked, overall they will be
        able to respond the same to the same congestion excursion.</t>

        <t>Therefore, for a congestion control to scale with packet size it
        should respond to dropped or marked bytes (as TFRC-SP <xref
        target="RFC4828"></xref> effectively does), not just to dropped or
        marked packets irrespective of packet size (as TCP does).</t>

        <t>The email <xref target="pktByteEmail"></xref> referred to by
        RFC2309 says the question of whether a packet's own size should affect
        its drop probability "depends on the dominant end-to-end congestion
        control mechanisms". But we argue the network layer should not be
        optimised for whatever transport is predominant.</t>

        <t>TCP congestion control ensures that flows competing for the same
        resource each maintain the same number of segments in flight,
        irrespective of segment size. So under similar conditions, flows with
        different segment sizes will get different bit rates. But even though
        reducing the drop probability of small packets helps ensure TCPs with
        different packet sizes will achieve similar bit rates, we argue this
        correction should be made to TCP itself, not to the network in order
        to fix one transport, no matter how prominent it is.</t>

        <t>Effectively, favouring small packets is reverse engineering of the
        network layer around TCP, contrary to the excellent advice in <xref
        target="RFC3426"></xref>, which asks designers to question "Why are
        you proposing a solution at this layer of the protocol stack, rather
        than at another layer?"</t>
      </section>

      <section anchor="pktb_Avoiding_Perverse_Incentives"
               title="Avoiding Perverse Incentives to (ab)use Smaller Packets">
        <t>Increasingly, it is being recognised that a protocol design must
        take care not to cause unintended consequences by giving the parties
        in the protocol exchange perverse incentives <xref
        target="Evol_cc"></xref><xref target="RFC3426"></xref>. Again, imagine
        a scenario where the same bit rate of packets will contribute the same
        to bit-congestion of a link irrespective of whether it is sent as
        fewer larger packets or more smaller packets. A protocol design that
        caused larger packets to be more likely to be dropped than smaller
        ones would be dangerous in this case:</t>

        <t><list style="hanging">
            <t hangText="Normal transports:">Even if a transport is not
            actually malicious, if it finds small packets go faster, over time
            it will tend to act in its own interest and use them. Queues that
            give advantage to small packets create an evolutionary pressure
            for transports to send at the same bit-rate but break their data
            stream down into tiny segments to reduce their drop rate.
            Encouraging a high volume of tiny packets might in turn
            unnecessarily overload a completely unrelated part of the system,
            perhaps more limited by header-processing than bandwidth.</t>

            <t hangText="Malicious transports:">A queue that gives an
            advantage to small packets can be used to amplify the force of a
            flooding attack. By sending a flood of small packets, the attacker
            can get the queue to discard more traffic in large packets,
            allowing more attack traffic to get through to cause further
            damage. Such a queue allows attack traffic to have a
            disproportionately large effect on regular traffic without the
            attacker having to do much work. <vspace blankLines="1" />Note
            that, although the byte-mode drop variant of RED amplifies small
            packet attacks, drop-tail queues amplify small packet attacks even
            more (see Security Considerations in <xref
            target="pktb_Security_Considerations"></xref>). Wherever possible
            neither should be used.</t>
          </list> Imagine two unresponsive flows arrive at a bit-congestible
        transmission link each with the same bit rate, say 1Mbps, but one
        consists of 1500B and the other 60B packets, which are 25x smaller.
        Consider a scenario where gentle RED <xref target="gentle_RED"></xref>
        is used, along with the variant of RED we advise against, i.e. where
        the RED algorithm is configured to adjust the drop probability of
        packets in proportion to each packet's size (byte mode packet drop).
        In this case, if RED drops 25% of the larger packets, it will aim to
        drop 1% of the smaller packets (but in practice it may drop more as
        congestion increases <xref target="RFC4828"></xref>(§B.4)<cref
        anchor="Note_Variation">The algorithm of the byte-mode drop variant of
        RED switches off any bias towards small packets whenever the smoothed
        queue length dictates that the drop probability of large packets
        should be 100%. In the example in the Introduction, as the large
        packet drop probability varies around 25% the small packet drop
        probability will vary around 1%, but with occasional jumps to 100%
        whenever the instantaneous queue (after drop) manages to sustain a
        length above the 100% drop point for longer than the queue averaging
        period.</cref>). Even though both flows arrive with the same bit rate,
        the bit rate the RED queue aims to pass to the line will be 750k for
        the flow of larger packet but 990k for the smaller packets (but
        because of rate variation it will be less than this target).</t>

        <t>It can be seen that this behaviour reopens the same denial of
        service vulnerability that drop tail queues offer to floods of small
        packet, though not necessarily as strongly (see <xref
        target="pktb_Security_Considerations"></xref>).</t>
      </section>

      <section anchor="pktb_Small.NE.Control" title="Small != Control">
        <t>It is tempting to drop small packets with lower probability to
        improve performance, because many control packets are small (TCP SYNs
        & ACKs, DNS queries & responses, SIP messages, HTTP GETs, etc)
        and dropping fewer control packets considerably improves performance.
        However, we must not give control packets preference purely by virtue
        of their smallness, otherwise it is too easy for any data source to
        get the same preferential treatment simply by sending data in smaller
        packets. Again we should not create perverse incentives to favour
        small packets rather than to favour control packets, which is what we
        intend.</t>

        <t>Just because many control packets are small does not mean all small
        packets are control packets.</t>

        <t>So again, rather than fix these problems in the network layer, we
        argue that the transport should be made more robust against losses of
        control packets (see 'Making Transports Robust against Control Packet
        Losses' in <xref
        target="pktb_Transport_Robust_Ctrl_Loss"></xref>).</t>
      </section>

      <section anchor="pktb_Impl_Efficiency" title="Implementation Efficiency">
        <t>Allowing for packet size at the transport rather than in the
        network ensures that neither the network nor the transport needs to do
        a multiply operation—multiplication by packet size is
        effectively achieved as a repeated add when the transport adds to its
        count of marked bytes as each congestion event is fed to it. This
        isn't a principled reason in itself, but it is a happy consequence of
        the other principled reasons.</t>
      </section>
    </section>

    <!--
    <section anchor="pktb_Measure" title="Congestion Measurement">
      <section anchor="pktb_Measurement_Q"
               title="Congestion Measurement by Queue Length">

        <t>Queue length is usually the most correct and simplest way to
        measure congestion of a resource. To avoid the pathological effects of
        drop tail, an AQM function can then be used to transform queue length
        into the probability of dropping or marking a packet (e.g. RED's
        piecewise linear function between thresholds). If the resource is
        bit-congestible, the length of the queue SHOULD be measured in bytes.
        If the resource is packet-congestible, the length of the queue SHOULD
        be measured in packets. No other choice makes sense, because the
        number of packets waiting in the queue isn't relevant if the resource
        gets congested by bytes and vice versa. We discuss the implications on
        RED's byte mode and packet mode for measuring queue length in <xref
        target="pktb_SotA" />.</t>

        <section anchor="pktb_Fixed_Buffers" title="Fixed Size Packet Buffers">
          <t>Some, mostly older, queuing hardware sets aside fixed sized
          buffers in which to store each packet in the queue. Also, with some
          hardware, any fixed sized buffers not completely filled by a packet
          are padded when transmitted to the wire. If we imagine a theoretical
          forwarding system with both queuing and transmission in fixed,
          MTU-sized units, it should clearly be treated as packet-congestible,
          because the queue length in packets would be a good model of
          congestion of the lower layer link.</t>

          <t>If we now imagine a hybrid forwarding system with transmission
          delay largely dependent on the byte-size of packets but buffers of
          one MTU per packet, it should strictly require a more complex
          algorithm to determine the probability of congestion. It should be
          treated as two resources in sequence, where the sum of the
          byte-sizes of the packets within each packet buffer models
          congestion of the line while the length of the queue in packets
          models congestion of the queue. Then the probability of congesting
          the forwarding buffer would be a conditional
          probability—conditional on the previously calculated
          probability of congesting the line.</t>

          <t>However, in systems that use fixed size buffers, it is unusual
          for all the buffers used by an interface to be the same size.
          Typically pools of different sized buffers are provided (Cisco uses
          the term 'buffer carving' for the process of dividing up memory into
          these pools <xref target="IOSArch" />). Usually, if the pool of
          small buffers is exhausted, arriving small packets can borrow space
          in the pool of large buffers, but not vice versa. However, it is
          easier to work out what should be done if we temporarily set aside
          the possibility of such borrowing. Then, with fixed pools of buffers
          for different sized packets and no borrowing, the size of each pool
          and the current queue length in each pool would both be measured in
          packets. So an AQM algorithm would have to maintain the queue length
          for each pool, and judge whether to drop/mark a packet of a
          particular size by looking at the pool for packets of that size and
          using the length (in packets) of its queue.</t>

          <t>We now return to the issue we temporarily set aside: small
          packets borrowing space in larger buffers. In this case, the only
          difference is that the pools for smaller packets have a maximum
          queue size that includes all the pools for larger packets. And every
          time a packet takes a larger buffer, the current queue size has to
          be incremented for all queues in the pools of buffers less than or
          equal to the buffer size used.</t>

          <t>We will return to borrowing of fixed sized buffers when we
          discuss biasing the drop/marking probability of a specific packet
          because of its size in <xref target="pktb_Network_Bias" />. But here
          we can give a simple summary of the present discussion on how to
          measure the length of queues of fixed buffers: no matter how
          complicated the scheme is, ultimately any fixed buffer system will
          need to measure its queue length in packets not bytes.</t>
        </section>
      </section>

      <section anchor="pktb_Measurement_NoQ"
               title="Congestion Measurement without a Queue">
        <t>AQM algorithms are nearly always described assuming there is a
        queue for a congested resource and the algorithm can use the queue
        length to determine the probability that it will drop or mark each
        packet. But not all congested resources lead to queues. For instance,
        wireless spectrum is bit-congestible (for a given coding scheme),
        because interference increases with the rate at which bits are
        transmitted. But wireless link protocols do not always maintain a
        queue that depends on spectrum interference. Similarly, power limited
        resources are also usually bit-congestible if energy is primarily
        required for transmission rather than header processing, but it is
        rare for a link protocol to build a queue as it approaches maximum
        power.</t>

        <t>However, AQM algorithms don't require a queue in order to work. For
        instance spectrum congestion can be modelled by signal quality using
        target bit-energy-to-noise-density ratio. And, to model radio power
        exhaustion, transmission power levels can be measured and compared to
        the maximum power available. <xref target="ECNFixedWireless" />
        proposes a practical and theoretically sound way to combine congestion
        notification for different bit-congestible resources at different
        layers along an end to end path, whether wireless or wired, and
        whether with or without queues.</t>
      </section>
    </section>

-->

    <!-- ================================================================ -->

    <section anchor="pktb_SotA" title="The State of the Art">
      <t>The original 1993 paper on RED <xref target="RED93"></xref> proposed
      two options for the RED active queue management algorithm: packet mode
      and byte mode. Packet mode measured the queue length in packets and
      dropped (or marked) individual packets with a probability independent of
      their size. Byte mode measured the queue length in bytes and marked an
      individual packet with probability in proportion to its size (relative
      to the maximum packet size). In the paper's outline of further work, it
      was stated that no recommendation had been made on whether the queue
      size should be measured in bytes or packets, but noted that the
      difference could be significant.</t>

      <t>When RED was recommended for general deployment in 1998 <xref
      target="RFC2309"></xref>, the two modes were mentioned implying the
      choice between them was a question of performance, referring to a 1997
      email <xref target="pktByteEmail"></xref> for advice on tuning. This
      email clarified that there were in fact two orthogonal choices: whether
      to measure queue length in bytes or packets (<xref
      target="pktb_Measure_Status"></xref> below) and whether the drop
      probability of an individual packet should depend on its own size (<xref
      target="pktb_Coding_Status"></xref> below).</t>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Measure_Status"
               title="Congestion Measurement: Status">
        <t>The choice of which metric to use to measure queue length was left
        open in RFC2309. It is now well understood that queues for
        bit-congestible resources should be measured in bytes, and queues for
        packet-congestible resources should be measured in packets.</t>

        <!-- (see <xref
        target="pktb_Measure" />).</t>
-->

        <t>Where buffers are not configured or legacy buffers cannot be
        configured to the above guideline, we do not have to make allowances
        for such legacy in future protocol design. If a bit-congestible buffer
        is measured in packets, the operator will have set the thresholds
        mindful of a typical mix of packets sizes. Any AQM algorithm on such a
        buffer will be oversensitive to high proportions of small packets,
        e.g. a DoS attack, and undersensitive to high proportions of large
        packets. But an operator can safely keep such a legacy buffer because
        any undersensitivity during unusual traffic mixes cannot lead to
        congestion collapse given the buffer will eventually revert to tail
        drop, discarding proportionately more large packets.</t>

        <t>Some modern queue implementations give a choice for setting RED's
        thresholds in byte-mode or packet-mode. This may merely be an
        administrator-interface preference, not altering how the queue itself
        is measured but on some hardware it does actually change the way it
        measures its queue. Whether a resource is bit-congestible or
        packet-congestible is a property of the resource, so an admin should
        not ever need to, or be able to, configure the way a queue measures
        itself.</t>

        <t>We believe the question of whether to measure queues in bytes or
        packets is fairly well understood these days. The only outstanding
        issues concern how to measure congestion when the queue is bit
        congestible but the resource is packet congestible or vice versa.
        There is no controversy over what should be done. It's just you have
        to be an expert in probability to work out what should be done
        (summarised in the following section) and, even if you have, it's not
        always easy to find a practical algorithm to implement it.</t>

        <section anchor="pktb_Fixed_Buffers" title="Fixed Size Packet Buffers">
          <t>Some, mostly older, queuing hardware sets aside fixed sized
          buffers in which to store each packet in the queue. Also, with some
          hardware, any fixed sized buffers not completely filled by a packet
          are padded when transmitted to the wire. If we imagine a theoretical
          forwarding system with both queuing and transmission in fixed,
          MTU-sized units, it should clearly be treated as packet-congestible,
          because the queue length in packets would be a good model of
          congestion of the lower layer link.</t>

          <t>If we now imagine a hybrid forwarding system with transmission
          delay largely dependent on the byte-size of packets but buffers of
          one MTU per packet, it should strictly require a more complex
          algorithm to determine the probability of congestion. It should be
          treated as two resources in sequence, where the sum of the
          byte-sizes of the packets within each packet buffer models
          congestion of the line while the length of the queue in packets
          models congestion of the queue. Then the probability of congesting
          the forwarding buffer would be a conditional
          probability—conditional on the previously calculated
          probability of congesting the line.</t>

          <t>In systems that use fixed size buffers, it is unusual for all the
          buffers used by an interface to be the same size. Typically pools of
          different sized buffers are provided (Cisco uses the term 'buffer
          carving' for the process of dividing up memory into these pools
          <xref target="IOSArch"></xref>). Usually, if the pool of small
          buffers is exhausted, arriving small packets can borrow space in the
          pool of large buffers, but not vice versa. However, it is easier to
          work out what should be done if we temporarily set aside the
          possibility of such borrowing. Then, with fixed pools of buffers for
          different sized packets and no borrowing, the size of each pool and
          the current queue length in each pool would both be measured in
          packets. So an AQM algorithm would have to maintain the queue length
          for each pool, and judge whether to drop/mark a packet of a
          particular size by looking at the pool for packets of that size and
          using the length (in packets) of its queue.</t>

          <t>We now return to the issue we temporarily set aside: small
          packets borrowing space in larger buffers. In this case, the only
          difference is that the pools for smaller packets have a maximum
          queue size that includes all the pools for larger packets. And every
          time a packet takes a larger buffer, the current queue size has to
          be incremented for all queues in the pools of buffers less than or
          equal to the buffer size used.</t>

          <t>We will return to borrowing of fixed sized buffers when we
          discuss biasing the drop/marking probability of a specific packet
          because of its size in <xref target="pktb_Network_Bias"></xref>. But
          here we can give a simple summary of the present discussion on how
          to measure the length of queues of fixed buffers: no matter how
          complicated the scheme is, ultimately any fixed buffer system will
          need to measure its queue length in packets not bytes.</t>
        </section>

        <section anchor="pktb_Measurement_NoQ"
                 title="Congestion Measurement without a Queue">
          <t>AQM algorithms are nearly always described assuming there is a
          queue for a congested resource and the algorithm can use the queue
          length to determine the probability that it will drop or mark each
          packet. But not all congested resources lead to queues. For
          instance, wireless spectrum is bit-congestible (for a given coding
          scheme), because interference increases with the rate at which bits
          are transmitted. But wireless link protocols do not always maintain
          a queue that depends on spectrum interference. Similarly, power
          limited resources are also usually bit-congestible if energy is
          primarily required for transmission rather than header processing,
          but it is rare for a link protocol to build a queue as it approaches
          maximum power.</t>

          <t>Nonetheless, AQM algorithms do not require a queue in order to
          work. For instance spectrum congestion can be modelled by signal
          quality using target bit-energy-to-noise-density ratio. And, to
          model radio power exhaustion, transmission power levels can be
          measured and compared to the maximum power available. <xref
          target="ECNFixedWireless"></xref> proposes a practical and
          theoretically sound way to combine congestion notification for
          different bit-congestible resources at different layers along an end
          to end path, whether wireless or wired, and whether with or without
          queues.</t>
        </section>
      </section>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Coding_Status" title="Congestion Coding: Status">
        <!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -  -->

        <section anchor="pktb_Network_Bias" title="Network Bias when Encoding">
          <t>The previously mentioned email <xref
          target="pktByteEmail"></xref> referred to by <xref
          target="RFC2309"></xref> gave advice we now disagree with. It said
          that drop probability should depend on the size of the packet being
          considered for drop if the resource is bit-congestible, but not if
          it is packet-congestible, but advised that most scarce resources in
          the Internet were currently bit-congestible. The argument continued
          that if packet drops were inflated by packet size (byte-mode
          dropping), "a flow's fraction of the packet drops is then a good
          indication of that flow's fraction of the link bandwidth in bits per
          second". This was consistent with a referenced policing mechanism
          being worked on at the time for detecting unusually high bandwidth
          flows, eventually published in 1999 <xref target="pBox"></xref>.
          [The problem could and should have been solved by making the
          policing mechanism count the volume of bytes randomly dropped, not
          the number of packets.]</t>

          <t>A few months before RFC2309 was published, an addendum was added
          to the above archived email referenced from the RFC, in which the
          final paragraph seemed to partially retract what had previously been
          said. It clarified that the question of whether the probability of
          dropping/marking a packet should depend on its size was not related
          to whether the resource itself was bit congestible, but a completely
          orthogonal question. However the only example given had the queue
          measured in packets but packet drop depended on the byte-size of the
          packet in question. No example was given the other way round.</t>

          <t>In 2000, Cnodder et al <xref target="REDbyte"></xref> pointed out
          that there was an error in the part of the original 1993 RED
          algorithm that aimed to distribute drops uniformly, because it
          didn't correctly take into account the adjustment for packet size.
          They recommended an algorithm called RED_4 to fix this. But they
          also recommended a further change, RED_5, to adjust drop rate
          dependent on the square of relative packet size. This was indeed
          consistent with one implied motivation behind RED's byte mode
          drop—that we should reverse engineer the network to improve
          the performance of dominant end-to-end congestion control
          mechanisms.</t>

          <t>By 2003, a further change had been made to the adjustment for
          packet size, this time in the RED algorithm of the ns2 simulator.
          Instead of taking each packet's size relative to a `maximum packet
          size' it was taken relative to a `mean packet size', intended to be
          a static value representative of the `typical' packet size on the
          link. We have not been able to find a justification for this change
          in the literature, however Eddy and Allman conducted experiments
          <xref target="REDbias"></xref> that assessed how sensitive RED was
          to this parameter, amongst other things. No-one seems to have
          pointed out that this changed algorithm can often lead to drop
          probabilities of greater than 1 [which should ring alarm bells
          hinting that there's a mistake in the theory somewhere]. On
          10-Nov-2004, this variant of byte-mode packet drop was made the
          default in the ns2 simulator.</t>

          <t>The byte-mode drop variant of RED is, of course, not the only
          possible bias towards small packets in queueing algorithms. We have
          already mentioned that tail-drop queues naturally tend to lock-out
          large packets once they are full. But also queues with fixed sized
          buffers reduce the probability that small packets will be dropped if
          (and only if) they allow small packets to borrow buffers from the
          pools for larger packets. As was explained in <xref
          target="pktb_Fixed_Buffers"></xref> on fixed size buffer carving,
          borrowing effectively makes the maximum queue size for small packets
          greater than that for large packets, because more buffers can be
          used by small packets while less will fit large packets.</t>

          <t>In itself, the bias towards small packets caused by buffer
          borrowing is perfectly correct. Lower drop probability for small
          packets is legitimate in buffer borrowing schemes, because small
          packets genuinely congest the machine's buffer memory less than
          large packets, given they can fit in more spaces. The bias towards
          small packets is not artificially added (as it is in RED's byte-mode
          drop algorithm), it merely reflects the reality of the way fixed
          buffer memory gets congested. Incidentally, the bias towards small
          packets from buffer borrowing is nothing like as large as that of
          RED's byte-mode drop.</t>

          <t>Nonetheless, fixed-buffer memory with tail drop is still prone to
          lock-out large packets, purely because of the tail-drop aspect. So a
          good AQM algorithm like RED with packet-mode drop should be used
          with fixed buffer memories where possible. If RED is too complicated
          to implement with multiple fixed buffer pools, the minimum necessary
          to prevent large packet lock-out is to ensure smaller packets never
          use the last available buffer in any of the pools for larger
          packets.</t>
        </section>

        <!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -  -->

        <section anchor="pktb_Transport_Bias"
                 title="Transport Bias when Decoding">
          <t>The above proposals to alter the network equipment to bias
          towards smaller packets have largely carried on outside the IETF
          process (unless one counts a reference in an informational RFC to an
          archived email!). Whereas, within the IETF, there are many different
          proposals to alter transport protocols to achieve the same goals,
          i.e. either to make the flow bit-rate take account of packet size,
          or to protect control packets from loss. This memo argues that
          altering transport protocols is the more principled approach.</t>

          <t>A recently approved experimental RFC adapts its transport layer
          protocol to take account of packet sizes relative to typical TCP
          packet sizes. This proposes a new small-packet variant of
          TCP-friendly rate control <xref target="RFC3448"></xref> called
          TFRC-SP <xref target="RFC4828"></xref>. Essentially, it proposes a
          rate equation that inflates the flow rate by the ratio of a typical
          TCP segment size (1500B including TCP header) over the actual
          segment size <xref target="PktSizeEquCC"></xref>. (There are also
          other important differences of detail relative to TFRC, such as
          using virtual packets <xref target="CCvarPktSize"></xref> to avoid
          responding to multiple losses per round trip and using a minimum
          inter-packet interval.)</t>

          <t>Section 4.5.1 of this TFRC-SP spec discusses the implications of
          operating in an environment where queues have been configured to
          drop smaller packets with proportionately lower probability than
          larger ones. But it only discusses TCP operating in such an
          environment, only mentioning TFRC-SP briefly when discussing how to
          define fairness with TCP. And it only discusses the byte-mode
          dropping version of RED as it was before Cnodder et al pointed out
          it didn't sufficiently bias towards small packets to make TCP
          independent of packet size.</t>

          <t>So the TFRC-SP spec doesn't address the issue of which of the
          network or the transport <spanx style="emph">should</spanx> handle
          fairness between different packet sizes. In its Appendix B.4 it
          discusses the possibility of both TFRC-SP and some network buffers
          duplicating each other's attempts to deliberately bias towards small
          packets. But the discussion is not conclusive, instead reporting
          simulations of many of the possibilities in order to assess
          performance but not recommending any particular course of
          action.</t>

          <t>The paper originally proposing TFRC with virtual packets
          (VP-TFRC) <xref target="CCvarPktSize"></xref> proposed that there
          should perhaps be two variants to cater for the different variants
          of RED. However, as the TFRC-SP authors point out, there is no way
          for a transport to know whether some queues on its path have
          deployed RED with byte-mode packet drop (except if an exhaustive
          survey found that no-one has deployed it!—see <xref
          target="pktb_Coding_Status_Summary"></xref>). Incidentally, VP-TFRC
          also proposed that byte-mode RED dropping should really square the
          packet size compensation factor (like that of RED_5, but apparently
          unaware of it).</t>

          <t>Pre-congestion notification <xref
          target="I-D.ietf-pcn"></xref> is a proposal to use
          a virtual queue for AQM marking for packets within one Diffserv
          class in order to give early warning prior to any real queuing. The
          proposed PCN marking algorithms have been designed not to take
          account of packet size when forwarding through queues. Instead the
          general principle has been to take account of the sizes of marked
          packets when monitoring the fraction of marking at the edge of the
          network.</t>
        </section>

        <!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -  -->

        <section anchor="pktb_Transport_Robust_Ctrl_Loss"
                 title="Making Transports Robust against Control Packet Losses">
          <t>Recently, two RFCs have defined changes to TCP that make it more
          robust against losing small control packets <xref
          target="RFC5562"></xref> <xref target="RFC5690"></xref>. In both
          cases they note that the case for these TCP changes would be weaker
          if RED were biased against dropping small packets. We argue here
          that these two proposals are a safer and more principled way to
          achieve TCP performance improvements than reverse engineering RED to
          benefit TCP.</t>

          <t>Although no proposals exist as far as we know, it would also be
          possible and perfectly valid to make control packets robust against
          drop by explicitly requesting a lower drop probability using their
          Diffserv code point <xref target="RFC2474"></xref> to request a
          scheduling class with lower drop.</t>

          <t>The re-ECN protocol proposal <xref
          target="I-D.briscoe-tsvwg-re-ecn-tcp"></xref> is designed so that
          transports can be made more robust against losing control packets.
          It gives queues an incentive to optionally give preference against
          drop to packets with the 'feedback not established' codepoint in the
          proposed 'extended ECN' field. Senders have incentives to use this
          codepoint sparingly, but they can use it on control packets to
          reduce their chance of being dropped. For instance, the proposed
          modification to TCP for re-ECN uses this codepoint on the SYN and
          SYN-ACK.</t>

          <t>Although not brought to the IETF, a simple proposal from Wischik
          <xref target="DupTCP"></xref> suggests that the first three packets
          of every TCP flow should be routinely duplicated after a short
          delay. It shows that this would greatly improve the chances of short
          flows completing quickly, but it would hardly increase traffic
          levels on the Internet, because Internet bytes have always been
          concentrated in the large flows. It further shows that the
          performance of many typical applications depends on completion of
          long serial chains of short messages. It argues that, given most of
          the value people get from the Internet is concentrated within short
          flows, this simple expedient would greatly increase the value of the
          best efforts Internet at minimal cost.</t>
        </section>

        <!-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -  -->

        <section anchor="pktb_Coding_Status_Summary"
                 title="Congestion Coding: Summary of Status">
          <?rfc needLines="6" ?>

          <texttable anchor="pktb_Tab_TFRC-SP"
                     title="Dependence of flow bit-rate per RTT on packet size s and drop rate p when network and/or transport bias towards small packets to varying degrees">
            <ttcol align="right">transport cc</ttcol>

            <ttcol align="center">RED_1 (packet mode drop)</ttcol>

            <ttcol align="center">RED_4 (linear byte mode drop)</ttcol>

            <ttcol align="center">RED_5 (square byte mode drop)</ttcol>

            <c>TCP or TFRC</c>

            <c>s/sqrt(p)</c>

            <c>sqrt(s/p)</c>

            <c>1/sqrt(p)</c>

            <c>TFRC-SP</c>

            <c>1/sqrt(p)</c>

            <c>1/sqrt(sp)</c>

            <c>1/(s.sqrt(p))</c>
          </texttable>

          <t><xref target="pktb_Tab_TFRC-SP"></xref> aims to summarise the
          positions we may now be in. Each column shows a different possible
          AQM behaviour in different queues in the network, using the
          terminology of Cnodder et al outlined earlier (RED_1 is basic RED
          with packet-mode drop). Each row shows a different transport
          behaviour: TCP <xref target="RFC5681"></xref> and TFRC <xref
          target="RFC3448"></xref> on the top row with TFRC-SP <xref
          target="RFC4828"></xref> below. Suppressing all inessential details
          the table shows that independence from packet size should either be
          achievable by not altering the TCP transport in a RED_5 network, or
          using the small packet TFRC-SP transport in a network without any
          byte-mode dropping RED (top right and bottom left). Top left is the
          `do nothing' scenario, while bottom right is the `do-both' scenario
          in which bit-rate would become far too biased towards small packets.
          Of course, if any form of byte-mode dropping RED has been deployed
          on a selection of congested queues, each path will present a
          different hybrid scenario to its transport.</t>

          <t>Whatever, we can see that the linear byte-mode drop column in the
          middle considerably complicates the Internet. It's a half-way house
          that doesn't bias enough towards small packets even if one believes
          the network should be doing the biasing. We argue below that <spanx
          style="emph">all</spanx> network layer bias towards small packets
          should be turned off—if indeed any equipment vendors have
          implemented it—leaving packet size bias solely as the preserve
          of the transport layer (solely the leftmost, packet-mode drop
          column).</t>

          <t>A survey has been conducted of 84 vendors to assess how widely
          drop probability based on packet size has been implemented in RED.
          Prior to the survey, an individual approach to Cisco received
          confirmation that, having checked the code-base for each of the
          product ranges, Cisco has not implemented any discrimination based
          on packet size in any AQM algorithm in any of its products. Also an
          individual approach to Alcatel-Lucent drew a confirmation that it
          was very likely that none of their products contained RED code that
          implemented any packet-size bias.</t>

          <t>Turning to our more formal survey (<xref
          target="pktb_Tab_RED_Survey"></xref>), about 19% of those surveyed
          have replied so far, giving a sample size of 16. Although we do not
          have permission to identify the respondents, we can say that those
          that have responded include most of the larger vendors, covering a
          large fraction of the market. They range across the large network
          equipment vendors at L3 & L2, firewall vendors, wireless
          equipment vendors, as well as large software businesses with a small
          selection of networking products. So far, all those who have
          responded have confirmed that they have not implemented the variant
          of RED with drop dependent on packet size (2 were fairly sure they
          had not but needed to check more thoroughly).</t>

          <texttable anchor="pktb_Tab_RED_Survey"
                     title="Vendor Survey on byte-mode drop variant of RED (lower drop probability for small packets)">
            <preamble></preamble>

            <ttcol align="right">Response</ttcol>

            <ttcol align="right">No. of vendors</ttcol>

            <ttcol align="right">%age of vendors</ttcol>

            <c>Not implemented</c>

            <c>14</c>

            <c>17%</c>

            <c>Not implemented (probably)</c>

            <c>2</c>

            <c>2%</c>

            <c>Implemented</c>

            <c>0</c>

            <c>0%</c>

            <c>No response</c>

            <c>68</c>

            <c>81%</c>

            <c>Total companies/orgs surveyed</c>

            <c>84</c>

            <c>100%</c>

            <postamble></postamble>
          </texttable>

          <t>Where reasons have been given, the extra complexity of packet
          bias code has been most prevalent, though one vendor had a more
          principled reason for avoiding it—similar to the argument of
          this document. We have established that Linux does not implement RED
          with packet size drop bias, although we have not investigated a
          wider range of open source code.</t>

          <t>Finally, we repeat that RED's byte mode drop is not the only way
          to bias towards small packets—tail-drop tends to lock-out
          large packets very effectively. Our survey was of vendor
          implementations, so we cannot be certain about operator deployment.
          But we believe many queues in the Internet are still tail-drop. The
          company of one of the co-authors (BT) has widely deployed RED, but
          there are bound to be many tail-drop queues, particularly in access
          network equipment and on middleboxes like firewalls, where RED is
          not always available.</t>

          <t>Routers using a memory architecture based on fixed size buffers
          with borrowing may also still be prevalent in the Internet. As
          explained in <xref target="pktb_Network_Bias"></xref>, these also
          provide a marginal (but legitimate) bias towards small packets. So
          even though RED byte-mode drop is not prevalent, it is likely there
          is still some bias towards small packets in the Internet due to tail
          drop and fixed buffer borrowing.</t>
        </section>
      </section>
    </section>

    <!-- ================================================================ -->

    <section anchor="pktb_Issues" title="Outstanding Issues and Next Steps">
      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Bit-World" title="Bit-congestible World">
        <t>For a connectionless network with nearly all resources being
        bit-congestible we believe the recommended position is now unarguably
        clear—that the network should not make allowance for packet
        sizes and the transport should. This leaves two outstanding issues:
        <list style="symbols">
            <t>How to handle any legacy of AQM with byte-mode drop already
            deployed;</t>

            <t>The need to start a programme to update transport congestion
            control protocol standards to take account of packet size.</t>
          </list></t>

        <t>The sample of returns from our vendor survey <xref
        target="pktb_Coding_Status_Summary"></xref> suggest that byte-mode
        packet drop seems not to be implemented at all let alone deployed, or
        if it is, it is likely to be very sparse. Therefore, we do not really
        need a migration strategy from all but nothing to nothing.</t>

        <t>A programme of standards updates to take account of packet size in
        transport congestion control protocols has started with TFRC-SP <xref
        target="RFC4828"></xref>, while weighted TCPs implemented in the
        research community <xref target="WindowPropFair"></xref> could form
        the basis of a future change to TCP congestion control <xref
        target="RFC5681"></xref> itself.</t>
      </section>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Bit-Pkt-World"
               title="Bit- & Packet-congestible World">
        <t>Nonetheless, a connectionless network with both bit-congestible and
        packet-congestible resources is a different matter. If we believe we
        should allow for this possibility in the future, this space contains a
        truly open research issue.</t>

        <t>We develop the concept of an idealised congestion notification 
protocol that supports both bit-congestible and packet-congestible resources 
in <xref target="pktb_Ideal"></xref>. The congestion notification requires at 
least two flags for
        congestion of bit-congestible and packet-congestible resources. This
        hides a fundamental problem—much more fundamental than whether
        we can magically create header space for yet another ECN flag in IPv4,
        or whether it would work while being deployed incrementally. A
        congestion notification protocol must survive a transition from low
        levels of congestion to high. Marking two states is feasible with
        explicit marking, but much harder if packets are dropped. Also, it
        will not always be cost-effective to implement AQM at every low level
        resource, so drop will often have to suffice. Distinguishing drop from
        delivery naturally provides just one congestion flag—it is hard
        to drop a packet in two ways that are distinguishable remotely. This
        is a similar problem to that of distinguishing wireless transmission
        losses from congestive losses.</t>

        <t>We should also note that, strictly, packet-congestible resources
        are actually cycle-congestible because load also depends on the
        complexity of each look-up and whether the pattern of arrivals is
        amenable to caching or not. Further, this reminds us that any solution
        must not require a forwarding engine to use excessive processor cycles
        in order to decide how to say it has no spare processor cycles.</t>

        <t>Recently, the dual resource queue (DRQ) proposal <xref
        target="DRQ"></xref> has been made on the premise that, as network
        processors become more cost effective, per packet operations will
        become more complex (irrespective of whether more function in the
        network layer is desirable). Consequently the premise is that CPU
        congestion will become more common. DRQ is a proposed modification to
        the RED algorithm that folds both bit congestion and packet congestion
        into one signal (either loss or ECN).</t>

        <t>The problem of signalling packet processing congestion is not
        pressing, as most Internet resources are designed to be
        bit-congestible before packet processing starts to congest (see <xref
        target="pktb_term"></xref>). However, the IRTF Internet congestion
        control research group (ICCRG) has set itself the task of reaching
        consensus on generic forwarding mechanisms that are necessary and
        sufficient to support the Internet's future congestion control
        requirements (the first challenge in <xref
        target="I-D.irtf-iccrg-welzl"></xref>).
        Therefore, rather than not giving this problem any thought at all,
        just because it is hard and currently hypothetical, we defer the
        question of whether packet congestion might become common and what to
        do if it does to the IRTF (the 'Small Packets' challenge in <xref
        target="I-D.irtf-iccrg-welzl"></xref>).</t>
      </section>
    </section>

    <!-- ================================================================ -->

    <section anchor="pktb_Conclusions" title="Recommendation and Conclusions">
      <t></t>

      <section anchor="pktb_Measure_Rec"
               title="Recommendation on Queue Measurement">
        <t>Queue length is usually the most correct and simplest way to
        measure congestion of a resource. To avoid the pathological effects of
        drop tail, an AQM function can then be used to transform queue length
        into the probability of dropping or marking a packet (e.g. RED's
        piecewise linear function between thresholds).</t>

        <t>If the resource is bit-congestible, the length of the queue SHOULD
        be measured in bytes. If the resource is packet-congestible, the
        length of the queue SHOULD be measured in packets. No other choice
        makes sense, because the number of packets waiting in the queue isn't
        relevant if the resource gets congested by bytes and vice versa. We
        discuss the implications on RED's byte mode and packet mode for
        measuring queue length in <xref target="pktb_SotA"></xref>.</t>

        <t>NOTE WELL that RED's byte-mode queue measurement is fine, being
        completely orthogonal to byte-mode drop. If a RED implementation has a
        byte-mode but does not specify what sort of byte-mode, it is most
        probably byte-mode queue measurement, which is fine. However, if in
        doubt, the vendor should be consulted.</t>
      </section>

      <section anchor="pktb_Notify_Rec"
               title="Recommendation on Notifying Congestion">
        <t>The strong recommendation is that AQM algorithms such as RED SHOULD
        NOT use byte-mode drop. More generally, the Internet's congestion
        notification protocols (drop, ECN & PCN) SHOULD take account of
        packet size when the notification is read by the transport layer, NOT
        when it is written by the network layer. This approach offers
        sufficient and correct congestion information for all known and future
        transport protocols and also ensures no perverse incentives are
        created that would encourage transports to use inappropriately small
        packet sizes.</t>

        <t>The alternative of deflating RED's drop probability for smaller
        packet sizes (byte-mode drop) has no enduring advantages. It is more
        complex, it creates the perverse incentive to fragment segments into
        tiny pieces and it reopens the vulnerability to floods of
        small-packets that drop-tail queues suffered from and AQM was designed
        to remove.</t>

        <t>Byte-mode drop is a change to the network layer that makes
        allowance for an omission from the design of TCP, effectively reverse
        engineering the network layer to contrive to make two TCPs with
        different packet sizes run at equal bit rates (rather than packet
        rates) under the same path conditions.</t>

        <t>It also improves TCP performance by reducing the chance that a SYN
        or a pure ACK will be dropped, because they are small. But we SHOULD
        NOT hack the network layer to improve or fix certain transport
        protocols. No matter how predominant a transport protocol is (even if
        it's TCP), trying to correct for its failings by biasing towards small
        packets in the network layer creates a perverse incentive to break
        down all flows from all transports into tiny segments.</t>

        <t>So far, our survey of 84 vendors across the industry has drawn
        responses from about 19%, none of whom have implemented the byte mode
        packet drop variant of RED. Given there appears to be little, if any,
        installed base it seems we can recommend removal of byte-mode drop
        from RED with little, if any, incremental deployment impact.</t>

        <t>If a vendor has implemented byte-mode drop, and an operator has
        turned it on, it is strongly RECOMMENDED that it SHOULD be turned off.
        Note that RED as a whole SHOULD NOT be turned off, as without it, a
        drop tail queue also biases against large packets. But note also that
        turning off byte-mode may alter the relative performance of
        applications using different packet sizes, so it would be advisable to
        establish the implications before turning it off.</t>
      </section>

      <section anchor="pktb_Respond_Rec"
               title="Recommendation on Responding to Congestion">
        <t>Instead of network equipment biasing its congestion notification
        for small packets, the IETF transport area should continue its
        programme of updating congestion control protocols to take account of
        packet size and to make transports less sensitive to losing control
        packets like SYNs and pure ACKS.</t>
      </section>

      <section anchor="pktb_Research_Rec" title="Recommended Future Research">
        <t>The above conclusions cater for the Internet as it is today with
        most, if not all, resources being primarily bit-congestible. A
        secondary conclusion of this memo is that we may see more
        packet-congestible resources in the future, so research may be needed
        to extend the Internet's congestion notification (drop or ECN) so that
        it can handle a mix of bit-congestible and packet-congestible
        resources.</t>
      </section>
    </section>

    <!-- ================================================================ -->

    <section anchor="pktb_Security_Considerations"
             title="Security Considerations">
      <t>This draft recommends that queues do not bias drop probability
      towards small packets as this creates a perverse incentive for
      transports to break down their flows into tiny segments. One of the
      benefits of implementing AQM was meant to be to remove this perverse
      incentive that drop-tail queues gave to small packets. Of course, if
      transports really want to make the greatest gains, they don't have to
      respond to congestion anyway. But we don't want applications that are
      trying to behave to discover that they can go faster by using smaller
      packets.</t>

      <t>In practice, transports cannot all be trusted to respond to
      congestion. So another reason for recommending that queues do not bias
      drop probability towards small packets is to avoid the vulnerability to
      small packet DDoS attacks that would otherwise result. One of the
      benefits of implementing AQM was meant to be to remove drop-tail's DoS
      vulnerability to small packets, so we shouldn't add it back again.</t>

      <t>If most queues implemented AQM with byte-mode drop, the resulting
      network would amplify the potency of a small packet DDoS attack. At the
      first queue the stream of packets would push aside a greater proportion
      of large packets, so more of the small packets would survive to attack
      the next queue. Thus a flood of small packets would continue on towards
      the destination, pushing regular traffic with large packets out of the
      way in one queue after the next, but suffering much less drop
      itself.</t>

      <t><xref target="pktb_Policing_Congestion_Response"></xref> explains why
      the ability of networks to police the response of <spanx style="emph">any</spanx>
      transport to congestion depends on bit-congestible network resources
      only doing packet-mode not byte-mode drop. In summary, it says that
      making drop probability depend on the size of the packets that bits
      happen to be divided into simply encourages the bits to be divided into
      smaller packets. Byte-mode drop would therefore irreversibly complicate
      any attempt to fix the Internet's incentive structures.</t>
    </section>

    <!-- ================================================================ -->

    <section anchor="pktb_Acknowledgements" title="Acknowledgements">
      <t>Thank you to Sally Floyd, who gave extensive and useful review
      comments. Also thanks for the reviews from Philip Eardley, Toby
      Moncaster and Arnaud Jacquet as well as helpful explanations of
      different hardware approaches from Larry Dunn and Fred Baker. I am
      grateful to Bruce Davie and his colleagues for providing a timely and
      efficient survey of RED implementation in Cisco's product range. Also
      grateful thanks to Toby Moncaster, Will Dormann, John Regnault, Simon
      Carter and Stefaan De Cnodder who further helped survey the current
      status of RED implementation and deployment and, finally, thanks to the
      anonymous individuals who responded.</t>

      <t>Bob Briscoe and Jukka Manner are partly funded by Trilogy, a research
      project (ICT- 216372) supported by the European Community under its
      Seventh Framework Programme. The views expressed here are those of the
      authors only.</t>
    </section>

    <!-- ================================================================ -->

    <section anchor="pktb_Comments_Solicited" title="Comments Solicited">
      <t>Comments and questions are encouraged and very welcome. They can be
      addressed to the IETF Transport Area working group mailing list
      <tsvwg@ietf.org>, and/or to the authors.</t>
    </section>
  </middle>

  <back>
    <!-- ================================================================ -->

    <references title="Normative References">
      <?rfc include="reference.RFC.2119" ?>

      <?rfc include="reference.RFC.2309" ?>

      <?rfc include="reference.RFC.3168" ?>

      <?rfc include="reference.RFC.3426" ?>

      <?rfc include='reference.RFC.5033'?>

    </references>

    <references title="Informative References">
      <?rfc include="localref.Floyd93.RED" ?>

      <?rfc include="localref.Floyd97.REDPktByteEmail" ?>

      <?rfc include="localref.Floyd99.Penalty_box" ?>

      <?rfc include="localref.Crowcroft98.MulTCP" ?>

      <?rfc include="localref.Gibbens99.Evol_cc" ?>

      <?rfc include="localref.Elloumi00.REDbyte" ?>

      <?rfc include="localref.Vasallo00.PktSizeEquCC" ?>

      <?rfc include="localref.Siris02a.Window_ECN" ?>

      <?rfc include="localref.Siris02.RscCtrlCDMA" ?>

      <?rfc include="reference.RFC.2474" ?>

      <?rfc include="reference.RFC.3714" ?>

      <?rfc include="reference.RFC.3448" ?>

      <?rfc include='reference.RFC.4828'?>

      <?rfc include="localref.Eddy03.REDbias" ?>

      <?rfc include="localref.Widmer04.CCvarPktSize" ?>

      <?rfc include="localref.I-D.briscoe-tsvwg-re-ecn-tcp" ?>

      <?rfc include="reference.I-D.ietf-pcn-marking-behaviour" ?>

      <?rfc include="localref.Briscoe07.Rate_fair_Dis" ?>

      <?rfc include="reference.RFC.5681" ?>

      <?rfc include="localref.I-D.falk-xcp-spec" ?>

      <?rfc include="reference.RFC.4782" ?>

      <?rfc include='localref.Floyd00.gentle_RED'?>
<!--
      <?rfc include='reference.I-D.ietf-tcpm-ecnsyn'?>

      <?rfc include='reference.I-D.floyd-tcpm-ackcc'?>
-->
      <?rfc include='localref.Wischik07.ShortMsgs'?>

      <?rfc include='localref.Shin08.DRQ'?>

      <?rfc include='localref.Bolla00.Cisco_IOS_Arch'?>

      <?rfc include='reference.I-D.irtf-iccrg-welzl-congestion-control-open-research'?>

      <?rfc include='reference.RFC.5670'?>

      <?rfc include='reference.RFC.5562'?>

      <?rfc include='reference.RFC.5690'?>

    </references>

    <!-- ================================================================ -->

    <!-- ================================================================ -->

    <section anchor="pktb_CN_Definition"
             title="Congestion Notification Definition: Further Justification">
      <t>In <xref target="pktb_term"></xref> on the definition of congestion
      notification, load not capacity was used as the denominator. This also
      has a subtle significance in the related debate over the design of new
      transport protocols—typical new protocol designs (e.g. in XCP
      <xref target="xcp-spec"></xref> & Quickstart <xref
      target="RFC4782"></xref>) expect the sending transport to communicate
      its desired flow rate to the network and network elements to
      progressively subtract from this so that the achievable flow rate
      emerges at the receiving transport.</t>

      <t>Congestion notification with total load in the denominator can serve
      a similar purpose (though in retrospect not in advance like XCP &
      QuickStart). Congestion notification is a dimensionless fraction but
      each source can extract necessary rate information from it because it
      already knows what its own rate is. Even though congestion notification
      doesn't communicate a rate explicitly, from each source's point of view
      congestion notification represents the fraction of the rate it was
      sending a round trip ago that couldn't (or wouldn't) be served by
      available resources. </t>
    </section>

<!-- ---------------------------------------------------- -->

      <!-- Old Section 5 ============================================ -->

<section anchor="pktb_Ideal" title="Idealised Wire Protocol">

        <t>We will start by inventing an idealised congestion notification
        protocol before discussing how to make it practical. The idealised
        protocol is shown to be correct using examples later in this appendix.
</t>

      <section anchor="pktb_Ideal_Coding"
               title="Protocol Coding">

        <t>Congestion notification involves the congested resource coding a
        congestion notification signal into the packet stream and the
        transports decoding it. The idealised protocol uses two different
        (imaginary) fields in each datagram to signal congestion: one for byte
        congestion and one for packet congestion.</t>

        <t>We are not saying two ECN fields will be needed (and we are not
        saying that somehow a resource should be able to drop a packet in one
        of two different ways so that the transport can distinguish which sort
        of drop it was!). These two congestion notification channels are just
        a conceptual device. They allow us to defer having to decide whether
        to distinguish between byte and packet congestion when the network
        resource codes the signal or when the transport decodes it.</t>

        <t>However, although this idealised mechanism isn't intended for
        implementation, we do want to emphasise that we may need to find a way
        to implement it, because it could become necessary to somehow
        distinguish between bit and packet congestion <xref
        target="RFC3714"></xref>. Currently, packet-congestion is not the
        common case, but there is no guarantee that it will not become common
        with future technology trends.</t>

        <t>The idealised wire protocol is given below. It accounts for packet
        sizes at the transport layer, not in the network, and then only in the
        case of bit-congestible resources. This avoids the perverse incentive
        to send smaller packets and the DoS vulnerability that would otherwise
        result if the network were to bias towards them (see the motivating
        argument about avoiding perverse incentives in <xref
        target="pktb_Avoiding_Perverse_Incentives"></xref>): <list
            style="numbers">
            <t>A packet-congestible resource trying to code congestion level
            p_p into a packet stream should mark the idealised `packet
            congestion' field in each packet with probability p_p irrespective
            of the packet's size. The transport should then take a packet with
            the packet congestion field marked to mean just one mark,
            irrespective of the packet size.</t>

            <t>A bit-congestible resource trying to code time-varying
            byte-congestion level p_b into a packet stream should mark the
            `byte congestion' field in each packet with probability p_b, again
            irrespective of the packet's size. Unlike before, the transport
            should take a packet with the byte congestion field marked to
            count as a mark on each byte in the packet.</t>
          </list></t>

        <t>The worked examples in <xref target="pktb_Scenarios"></xref> show
        that transports can extract sufficient and correct congestion
        notification from these protocols for cases when two flows with
        different packet sizes have matching bit rates or matching packet
        rates. Examples are also given that mix these two flows into one to
        show that a flow with mixed packet sizes would still be able to
        extract sufficient and correct information.</t>

        <t>Sufficient and correct congestion information means that there is
        sufficient information for the two different types of transport
        requirements: <list style="hanging">
            <t hangText="Ratio-based:">Established transport congestion
            controls like TCP's <xref target="RFC5681"></xref> aim to achieve
            equal segment rates per RTT through the same bottleneck—TCP
            friendliness <xref target="RFC3448"></xref>. They work with the
            ratio of dropped to delivered segments (or marked to unmarked
            segments in the case of ECN). The example scenarios show that
            these ratio-based transports are effectively the same whether
            counting in bytes or packets, because the units cancel out.
            (Incidentally, this is why TCP's bit rate is still proportional to
            packet size even when byte-counting is used, as recommended for
            TCP in <xref target="RFC5681"></xref>, mainly for orthogonal
            security reasons.)</t>

            <t hangText="Absolute-target-based:">Other congestion controls
            proposed in the research community aim to limit the volume of
            congestion caused to a constant weight parameter. <xref
            target="MulTCP"></xref><xref target="WindowPropFair"></xref> are
            examples of weighted proportionally fair transports designed for
            cost-fair environments <xref target="Rate_fair_Dis"></xref>. In
            this case, the transport requires a count (not a ratio) of
            dropped/marked bytes in the bit-congestible case and of
            dropped/marked packets in the packet congestible case.</t>
          </list></t>
</section>

    <section anchor="pktb_Scenarios" title="Example Scenarios">
      <!--{ToDo: Tabulate these subsections}-->

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Notation" title="Notation">
        <t>To prove our idealised wire protocol (<xref
        target="pktb_Ideal_Coding"></xref>) is correct, we will compare two
        flows with different packet sizes, s_1 and s_2 [bit/pkt], to make sure
        their transports each see the correct congestion notification.
        Initially, within each flow we will take all packets as having equal
        sizes, but later we will generalise to flows within which packet sizes
        vary. A flow's bit rate, x [bit/s], is related to its packet rate, u
        [pkt/s], by <list style="empty">
            <t>x(t) = s.u(t).</t>
          </list></t>

        <t>We will consider a 2x2 matrix of four scenarios:</t>

        <?rfc needLines="6" ?>

        <texttable anchor="pktb_Tab_Scenarios">
          <ttcol align="right">resource type and congestion level</ttcol>

          <ttcol align="center">A) Equal bit rates</ttcol>

          <ttcol align="center">B) Equal pkt rates</ttcol>

          <c>i) bit-congestible, p_b</c>

          <c>(Ai)</c>

          <c>(Bi)</c>

          <c>ii) pkt-congestible, p_p</c>

          <c>(Aii)</c>

          <c>(Bii)</c>
        </texttable>
      </section>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Ai"
               title="Bit-congestible resource, equal bit rates (Ai)">
        <t>Starting with the bit-congestible scenario, for two flows to
        maintain equal bit rates (Ai) the ratio of the packet rates must be
        the inverse of the ratio of packet sizes: u_2/u_1 = s_1/s_2. So, for
        instance, a flow of 60B packets would have to send 25x more packets to
        achieve the same bit rate as a flow of 1500B packets. If a congested
        resource marks proportion p_b of packets irrespective of size, the
        ratio of marked packets received by each transport will still be the
        same as the ratio of their packet rates, p_b.u_2/p_b.u_1 = s_1/s_2. So
        of the 25x more 60B packets sent, 25x more will be marked than in the
        1500B packet flow, but 25x more won't be marked too.</t>

        <t>In this scenario, the resource is bit-congestible, so it always
        uses our idealised bit-congestion field when it marks packets.
        Therefore the transport should count marked bytes not packets. But it
        doesn't actually matter for ratio-based transports like TCP (<xref
        target="pktb_Ideal_Coding"></xref>). The ratio of marked to unmarked
        bytes seen by each flow will be p_b, as will the ratio of marked to
        unmarked packets. Because they are ratios, the units cancel out.</t>

        <t>If a flow sent an inconsistent mixture of packet sizes, we have
        said it should count the ratio of marked and unmarked bytes not
        packets in order to correctly decode the level of congestion. But
        actually, if all it is trying to do is decode p_b, it still doesn't
        matter. For instance, imagine the two equal bit rate flows were
        actually one flow at twice the bit rate sending a mixture of one 1500B
        packet for every thirty 60B packets. 25x more small packets will be
        marked and 25x more will be unmarked. The transport can still
        calculate p_b whether it uses bytes or packets for the ratio. In
        general, for any algorithm which works on a ratio of marks to
        non-marks, either bytes or packets can be counted interchangeably,
        because the choice cancels out in the ratio calculation.</t>

        <t>However, where an absolute target rather than relative volume of
        congestion caused is important (<xref
        target="pktb_Ideal_Coding"></xref>), as it is for congestion
        accountability <xref target="Rate_fair_Dis"></xref>, the transport
        must count marked bytes not packets, in this bit-congestible case.
        Aside from the goal of congestion accountability, this is how the bit
        rate of a transport can be made independent of packet size; by
        ensuring the rate of congestion caused is kept to a constant weight
        <xref target="WindowPropFair"></xref>, rather than merely responding
        to the ratio of marked and unmarked bytes.</t>

        <t>Note the unit of byte-congestion-volume is the byte.</t>
      </section>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Bi"
               title="Bit-congestible resource, equal packet rates (Bi)">
        <t>If two flows send different packet sizes but at the same packet
        rate, their bit rates will be in the same ratio as their packet sizes,
        x_2/x_1 = s_2/s_1. For instance, a flow sending 1500B packets at the
        same packet rate as another sending 60B packets will be sending at 25x
        greater bit rate. In this case, if a congested resource marks
        proportion p_b of packets irrespective of size, the ratio of packets
        received with the byte-congestion field marked by each transport will
        be the same, p_b.u_2/p_b.u_1 = 1.</t>

        <t>Because the byte-congestion field is marked, the transport should
        count marked bytes not packets. But because each flow sends
        consistently sized packets it still doesn't matter for ratio-based
        transports. The ratio of marked to unmarked bytes seen by each flow
        will be p_b, as will the ratio of marked to unmarked packets.
        Therefore, if the congestion control algorithm is only concerned with
        the ratio of marked to unmarked packets (as is TCP), both flows will
        be able to decode p_b correctly whether they count packets or
        bytes.</t>

        <t>But if the absolute volume of congestion is important, e.g. for
        congestion accountability, the transport must count marked bytes not
        packets. Then the lower bit rate flow using smaller packets will
        rightly be perceived as causing less byte-congestion even though its
        packet rate is the same.</t>

        <t>If the two flows are mixed into one, of bit rate x1+x2, with equal
        packet rates of each size packet, the ratio p_b will still be
        measurable by counting the ratio of marked to unmarked bytes (or
        packets because the ratio cancels out the units). However, if the
        absolute volume of congestion is required, the transport must count
        the sum of congestion marked bytes, which indeed gives a correct
        measure of the rate of byte-congestion p_b(x_1 + x_2) caused by the
        combined bit rate.</t>
      </section>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Aii"
               title="Pkt-congestible resource, equal bit rates (Aii)">
        <t>Moving to the case of packet-congestible resources, we now take two
        flows that send different packet sizes at the same bit rate, but this
        time the pkt-congestion field is marked by the resource with
        probability p_p. As in scenario Ai with the same bit rates but a
        bit-congestible resource, the flow with smaller packets will have a
        higher packet rate, so more packets will be both marked and unmarked,
        but in the same proportion.</t>

        <t>This time, the transport should only count marks without taking
        into account packet sizes. Transports will get the same result, p_p,
        by decoding the ratio of marked to unmarked packets in either
        flow.</t>

        <t>If one flow imitates the two flows but merged together, the bit
        rate will double with more small packets than large. The ratio of
        marked to unmarked packets will still be p_p. But if the absolute
        number of pkt-congestion marked packets is counted it will accumulate
        at the combined packet rate times the marking probability,
        p_p(u_1+u_2), 26x faster than packet congestion accumulates in the
        single 1500B packet flow of our example, as required.</t>

        <t>But if the transport is interested in the absolute number of packet
        congestion, it should just count how many marked packets arrive. For
        instance, a flow sending 60B packets will see 25x more marked packets
        than one sending 1500B packets at the same bit rate, because it is
        sending more packets through a packet-congestible resource.</t>

        <t>Note the unit of packet congestion is a packet.</t>
      </section>

      <!-- ________________________________________________________________ -->

      <section anchor="pktb_Bii"
               title="Pkt-congestible resource, equal packet rates (Bii)">
        <t>Finally, if two flows with the same packet rate, pass through a
        packet-congestible resource, they will both suffer the same proportion
        of marking, p_p, irrespective of their packet sizes. On detecting that
        the pkt-congestion field is marked, the transport should count
        packets, and it will be able to extract the ratio p_p of marked to
        unmarked packets from both flows, irrespective of packet sizes.</t>

        <t>Even if the transport is monitoring the absolute amount of packets
        congestion over a period, still it will see the same amount of packet
        congestion from either flow.</t>

        <t>And if the two equal packet rates of different size packets are
        mixed together in one flow, the packet rate will double, so the
        absolute volume of packet-congestion will accumulate at twice the rate
        of either flow, 2p_p.u_1 = p_p(u_1+u_2).</t>
      </section>

    </section>

</section>

<!------------------------------------------------------------->
    <section anchor="pktb_Policing_Congestion_Response"
             title="Byte-mode Drop Complicates Policing Congestion Response">
      <t>This appendix explains why the ability of networks to police the
      response of <spanx style="emph">any</spanx> transport to congestion
      depends on bit-congestible network resources only doing packet-mode not
      byte-mode drop.</t>

      <t>To be able to police a transport's response to congestion when
      fairness can only be judged over time and over all an individual's
      flows, the policer has to have an integrated view of all the congestion
      an individual (not just one flow) has caused due to all traffic entering
      the Internet from that individual. This is termed congestion
      accountability.</t>

      <t>But a byte-mode drop algorithm has to depend on the local MTU of the
      line - an algorithm needs to use some concept of a 'normal' packet size.
      Therefore, one dropped or marked packet is not necessarily equivalent to
      another unless you know the MTU at the queue where it was
      dropped/marked. To have an integrated view of a user, we believe
      congestion policing has to be located at an individual's attachment
      point to the Internet <xref
      target="I-D.briscoe-tsvwg-re-ecn-tcp"></xref>. But from there it cannot
      know the MTU of each remote queue that caused each drop/mark. Therefore
      it cannot take an integrated approach to policing all the responses to
      congestion of all the transports of one individual. Therefore it cannot
      police anything.</t>

      <t>The security/incentive argument <spanx style="emph">for</spanx>
      packet-mode drop is similar. Firstly, confining RED to packet-mode drop
      would not preclude bottleneck policing approaches such as <xref
      target="pBox"></xref> as it seems likely they could work just as well by
      monitoring the volume of dropped bytes rather than packets. Secondly
      packet-mode dropping/marking naturally allows the congestion
      notification of packets to be globally meaningful without relying on MTU
      information held elsewhere.</t>

      <t>Because we recommend that a dropped/marked packet should be taken to
      mean that all the bytes in the packet are dropped/marked, a policer can
      remain robust against bits being re-divided into different size packets
      or across different size flows <xref target="Rate_fair_Dis"></xref>.
      Therefore policing would work naturally with just simple packet-mode
      drop in RED.</t>

      <t>In summary, making drop probability depend on the size of the packets
      that bits happen to be divided into simply encourages the bits to be
      divided into smaller packets. Byte-mode drop would therefore
      irreversibly complicate any attempt to fix the Internet's incentive
      structures.</t>
    </section>

    <section anchor="changelog" title="Changes from Previous Versions">
      <t>To be removed by the RFC Editor on publication.</t>

      <t>Full incremental diffs between each version are available at
      <http://www.cs.ucl.ac.uk/staff/B.Briscoe/pubs.html#byte-pkt-congest>
      or
      <http://tools.ietf.org/wg/tsvwg/draft-ietf-tsvwg-byte-pkt-congest/>
      (courtesy of the rfcdiff tool): <list style="hanging">
          <t hangText="From -01 to -02 (this version):"><list style="symbols">
              <t>Restructured the whole document for (hopefully) easier
              reading and clarity. The concrete recommendation, in RFC2119
              language, is now in <xref target="pktb_Conclusions"></xref>.</t>
            </list></t>

          <t hangText="From -00 to -01:"><list style="symbols">
              <t>Minor clarifications throughout and updated references</t>
            </list></t>

          <t
          hangText="From briscoe-byte-pkt-mark-02 to ietf-byte-pkt-congest-00:"><list
              style="symbols">
              <t>Added note on relationship to existing RFCs</t>

              <t>Posed the question of whether packet-congestion could become
              common and deferred it to the IRTF ICCRG. Added ref to the
              dual-resource queue (DRQ) proposal.</t>

              <t>Changed PCN references from the PCN charter &
              architecture to the PCN marking behaviour draft most likely to
              imminently become the standards track WG item.</t>
            </list></t>

          <t hangText="From -01 to -02:"><list style="symbols">
              <t>Abstract reorganised to align with clearer separation of
              issue in the memo.</t>

              <t>Introduction reorganised with motivating arguments removed to
              new <xref target="pktb_Motivation"></xref>.</t>

              <t>Clarified avoiding lock-out of large packets is not the main
              or only motivation for RED.</t>

              <t>Mentioned choice of drop or marking explicitly throughout,
              rather than trying to coin a word to mean either.</t>

              <t>Generalised the discussion throughout to any packet
              forwarding function on any network equipment, not just
              routers.</t>

              <t>Clarified the last point about why this is a good time to
              sort out this issue: because it will be hard / impossible to
              design new transports unless we decide whether the network or
              the transport is allowing for packet size.</t>

              <t>Added statement explaining the horizon of the memo is long
              term, but with short term expediency in mind.</t>

              <t>Added material on scaling congestion control with packet size
              (<xref target="pktb_Scaling"></xref>).</t>

              <t>Separated out issue of normalising TCP's bit rate from issue
              of preference to control packets (<xref
              target="pktb_Small.NE.Control"></xref>).</t>

              <t>Divided up Congestion Measurement section for clarity,
              including new material on fixed size packet buffers and buffer
              carving (<xref target="pktb_Fixed_Buffers"></xref> & <xref
              target="pktb_Network_Bias"></xref>) and on congestion
              measurement in wireless link technologies without queues (<xref
              target="pktb_Measurement_NoQ"></xref>).</t>

              <t>Added section on 'Making Transports Robust against Control
              Packet Losses' (<xref
              target="pktb_Transport_Robust_Ctrl_Loss"></xref>) with existing
              & new material included.</t>

              <t>Added tabulated results of vendor survey on byte-mode drop
              variant of RED (<xref target="pktb_Tab_RED_Survey"></xref>).</t>
            </list></t>

          <t hangText="From -00 to -01:"><list style="symbols">
              <t>Clarified applicability to drop as well as ECN.</t>

              <t>Highlighted DoS vulnerability.</t>

              <t>Emphasised that drop-tail suffers from similar problems to
              byte-mode drop, so only byte-mode drop should be turned off, not
              RED itself.</t>

              <t>Clarified the original apparent motivations for recommending
              byte-mode drop included protecting SYNs and pure ACKs more than
              equalising the bit rates of TCPs with different segment sizes.
              Removed some conjectured motivations.</t>

              <t>Added support for updates to TCP in progress (ackcc &
              ecn-syn-ack).</t>

              <t>Updated survey results with newly arrived data.</t>

              <t>Pulled all recommendations together into the conclusions.</t>

              <t>Moved some detailed points into two additional appendices and
              a note.</t>

              <t>Considerable clarifications throughout.</t>

              <t>Updated references</t>
            </list></t>
        </list></t>
    </section>
  </back>
</rfc>

PAFTECH AB 2003-20262026-04-23 12:06:09