Identifying Modified Explicit Congestion Notification (ECN) Semantics for Ultra-Low Queuing Delay
draft-ietf-tsvwg-ecn-l4s-id-00

Abstract

This specification defines the identifier to be used on IP packets for a new network service called low latency, low loss and scalable throughput (L4S). It is similar to the original (or 'Classic') Explicit Congestion Notification (ECN). 'Classic' ECN marking was required to be equivalent to a drop, both when applied in the network and when responded to by a transport. Unlike 'Classic' ECN marking, for packets carrying the L4S identifier, the network applies marking more immediately and more aggressively than drop, and the transport response to each mark is reduced and smoothed relative to that for drop. The two changes counterbalance each other so that the throughput of an L4S flow will be roughly the same as a 'Classic' flow under the same conditions. However, the much more frequent control signals and the finer responses to them result in ultra-low queuing delay without compromising link utilization, even during high load. Examples of new active queue management (AQM) marking algorithms and examples of new transports (whether TCP-like or real-time) are specified separately. The new L4S identifier is the key piece that enables them to interwork and distinguishes them from 'Classic' traffic. It gives an incremental migration path so that existing 'Classic' TCP traffic will be no worse off, but it can be prevented from degrading the ultra-low delay and loss of the new scalable transports.

Status of This Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at http://datatracker.ietf.org/drafts/current/.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."

This Internet-Draft will expire on November 6, 2017.

Copyright Notice

Copyright (c) 2017 IETF Trust and the persons identified as the document authors. All rights reserved.

This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.

Table of Contents

• 1. Introduction
• 1.1. Problem
• 1.2. Terminology
• 1.3. Scope
• 2. L4S Packet Identifier
• 2.1. Consensus Choice of L4S Packet Identifier: Requirements
• 2.2. L4S Packet Identification at Run-Time
• 2.3. Pre-Requisite Transport Layer Behaviour
• 2.3.1. Pre-Requisite Congestion Response
• 2.3.2. Pre-Requisite Transport Feedback
• 2.4. Exception for L4S Packet Identification by Network Nodes with Transport-Layer Awareness
• 2.5. The Meaning of L4S CE Relative to Drop
• 3. IANA Considerations
• 4. Security Considerations
• 5. Acknowledgements
• 6. References
• 6.1. Normative References
• 6.2. Informative References
• Appendix A. The 'TCP Prague Requirements'
• A.1. Requirements for Scalable Transport Protocols
• A.1.1. Use of L4S Packet Identifier
• A.1.2. Accurate ECN Feedback
• A.1.3. Fall back to Reno/Cubic congestion control on packet loss
• A.1.4. Fall back to Reno/Cubic congestion control on classic ECN bottlenecks
• A.1.5. Reduce RTT dependence
• A.1.6. Scaling down the congestion window
• A.2. Scalable Transport Protocol Optimizations
• A.2.1. Setting ECT in TCP Control Packets and Retransmissions
• A.2.2. Faster than Additive Increase
• A.2.3. Faster Convergence at Flow Start
• Appendix B. Alternative Identifiers
• B.1. ECT(1) and CE codepoints
• B.2. ECN Plus a Diffserv Codepoint (DSCP)
• B.3. ECN capability alone
• B.4. Protocol ID
• B.5. Source or destination addressing
• B.6. Summary: Merits of Alternative Identifiers
• Appendix C. Potential Competing Uses for the ECT(1) Codepoint
• C.1. Integrity of Congestion Feedback
• C.2. Notification of Less Severe Congestion than CE
• Authors' Addresses

1. Introduction

This specification defines the identifier to be used on IP packets for a new network service called low latency, low loss and scalable throughput (L4S). It is similar to the original (or 'Classic') Explicit Congestion Notification (ECN). 'Classic' ECN marking was required to be equivalent to a drop, both when applied in the network and when responded to by a transport. Unlike 'Classic' ECN marking, the network applies L4S marking more immediately and more aggressively than drop, and the transport response to each mark is reduced and smoothed relative to that for drop. The two changes counterbalance each other so that the bit-rate of an L4S flow will be roughly the same as a 'Classic' flow under the same conditions. Nonetheless, the much more frequent control signals and the finer responses to them result in ultra-low queuing delay without compromising link utilization, even during high load.

An example of an active queue management (AQM) marking algorithm that enables the L4S service is the DualQ Coupled AQM defined in a complementary specification [I-D.ietf-tsvwg-aqm-dualq-coupled]. An example of a scalable transport that would enable the L4S service is Data Centre TCP (DCTCP), which until now has been applicable solely to controlled environments like data centres [I-D.ietf-tcpm-dctcp], because it is too aggressive to co-exist with existing TCP. However, AQMs like DualQ Coupled enable scalable transports like DCTCP to co-exist with existing traffic, each getting roughly the same flow rate when they compete under similar conditions. Note that DCTCP will still not be safe to deploy on the Internet until it satisfies the requirements listed in Section 2.3.

The new L4S identifier is the key piece that enables these two parts to interwork and distinguishes them from 'Classic' traffic. It gives an incremental migration path so that existing 'Classic' TCP traffic will be no worse off, but it can be prevented from degrading the ultra-low delay and loss of the new scalable transports. The performance improvement is so great that it is hoped it will motivate initial deployment of the separate parts of this system.

1.1. Problem

Latency is becoming the critical performance factor for many (most?) applications on the public Internet, e.g. Web, voice, conversational video, gaming, finance apps, remote desktop and cloud-based applications. In the developed world, further increases in access network bit-rate offer diminishing returns, whereas latency is still a multi-faceted problem. In the last decade or so, much has been done to reduce propagation time by placing caches or servers closer to users. However, queuing remains a major component of latency.

The Diffserv architecture provides Expedited Forwarding [RFC3246], so that low latency traffic can jump the queue of other traffic. However, on access links dedicated to individual sites (homes, small enterprises or mobile devices), often all traffic at any one time will be latency-sensitive. Then Diffserv is of little use. Instead, we need to remove the causes of any unnecessary delay.

The bufferbloat project has shown that excessively-large buffering ('bufferbloat') has been introducing significantly more delay than the underlying propagation time. These delays appear only intermittently—only when a capacity-seeking (e.g. TCP) flow is long enough for the queue to fill the buffer, making every packet in other flows sharing the buffer sit through the queue.

Active queue management (AQM) was originally developed to solve this problem (and others). Unlike Diffserv, which gives low latency to some traffic at the expense of others, AQM controls latency for all traffic in a class. In general, AQMs introduce an increasing level of discard from the buffer the longer the queue persists above a shallow threshold. This gives sufficient signals to capacity-seeking (aka. greedy) flows to keep the buffer empty for its intended purpose: absorbing bursts. However, RED [RFC2309] and other algorithms from the 1990s were sensitive to their configuration and hard to set correctly. So, AQM was not widely deployed.

More recent state-of-the-art AQMs, e.g. fq_CoDel [I-D.ietf-aqm-fq-codel], PIE [I-D.ietf-aqm-pie], Adaptive RED [ARED01], are easier to configure, because they define the queuing threshold in time not bytes, so it is invariant for different link rates. However, no matter how good the AQM, the sawtoothing rate of TCP will either cause queuing delay to vary or cause the link to be under-utilized. Even with a perfectly tuned AQM, the additional queuing delay will be of the same order as the underlying speed-of-light delay across the network. Flow-queuing can isolate one flow from another, but it cannot isolate a TCP flow from the delay variations it inflicts on itself, and it has other problems - it overrides the flow rate decisions of variable rate video applications, it does not recognise the flows within IPSec VPN tunnels and it is relatively expensive to implement.

Latency is not our only concern: It was known when TCP was first developed that it would not scale to high bandwidth-delay products. Given regular broadband bit-rates over WAN distances are already [RFC3649] beyond the scaling range of 'Classic' TCP Reno, 'less unscalable' Cubic [I-D.ietf-tcpm-cubic] and Compound [I-D.sridharan-tcpm-ctcp] variants of TCP have been successfully deployed. However, these are now approaching their scaling limits. Unfortunately, fully scalable TCPs such as DCTCP [I-D.ietf-tcpm-dctcp] cause 'Classic' TCP to starve itself, which is why they have been confined to private data centres or research testbeds (until now).

It turns out that a TCP algorithm like DCTCP that solves TCP's scalability problem also solves the latency problem, because the finer sawteeth cause very little queuing delay. A supporting paper [DCttH15] gives the full explanation of why the design solves both the latency and the scaling problems, both in plain English and in more precise mathematical form. The explanation is summarised without the maths in the L4S architecture document [I-D.ietf-tsvwg-l4s-arch].

1.2. Terminology

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119]. In this document, these words will appear with that interpretation only when in ALL CAPS. Lower case uses of these words are not to be interpreted as carrying RFC-2119 significance.

Classic service:

The 'Classic' service is intended for all the behaviours that currently co-exist with TCP Reno (e.g. TCP Cubic, Compound, SCTP, etc).

Low-Latency, Low-Loss and Scalable (L4S) service:

The 'L4S' service is intended for traffic from scalable TCP algorithms such as Data Centre TCP. But it is also more general—it will allow a set of congestion controls with similar scaling properties to DCTCP (e.g. Relentless [Mathis09]) to evolve.

Both Classic and L4S services can cope with a proportion of unresponsive or less-responsive traffic as well (e.g. DNS, VoIP, etc).

Classic ECN:

The original Explicit Congestion Notification (ECN) protocol [RFC3168].

1.3. Scope

The new L4S identifier defined in this specification is applicable for IPv4 and IPv6 packets (as for classic ECN [RFC3168]). It is applicable for the unicast, multicast and anycast forwarding modes. It is an orthogonal packet classification to Differentiated Services (Diffserv [RFC2474]), therefore it can be applied to any packet in any Diffserv traffic class. However, as with classic ECN, any particular forwarding node might not implement an active queue management algorithm in all its Diffserv queues.

This document is intended for experimental status, so it does not update any standards track RFCs. Therefore it depends on [I-D.ietf-tsvwg-ecn-experimentation], which:

• updates the ECN proposed standard [RFC3168] (in certain specified cases including the present document) to relax the requirement that an ECN mark must be equivalent to a drop, both when applied by the network, and when responded to by the sender;
• makes the case for changing the status of the experimental ECN nonce [RFC3540] to historic (see Appendix C.1 for rationale);
• makes consequent updates to the following proposed standard RFCs to reflect the above two bullets:
  • ECN for RTP [RFC6679];
  • the congestion control specifications of various DCCP congestion control identifier (CCID) profiles [RFC4341], [RFC4342], [RFC5622].

2. L4S Packet Identifier

2.1. Consensus Choice of L4S Packet Identifier: Requirements

This subsection briefly records the process that led to a consensus choice of L4S identifier, selected from all the alternatives in Appendix B.

Ideally, the identifier for packets using the Low Latency, Low Loss, Scalable throughput (L4S) service ought to meet the following requirements:

• it SHOULD survive end-to-end between source and destination applications: across the boundary between host and network, between interconnected networks, and through middleboxes;
• it SHOULD be common to IPv4 and IPv6 and transport agnostic;
• it SHOULD be incrementally deployable;
• it SHOULD enable an AQM to classify packets encapsulated by outer IP or lower-layer headers;
• it SHOULD consume minimal extra codepoints;
• it SHOULD not lead to some packets of a transport-layer flow being served by a different queue from others.

Whether the identifier would be recoverable if the experiment failed is a factor that could be taken into account. However, this has not been made a requirement, because that would favour schemes that would be easier to fail, rather than those more likely to succeed.

It is recognised that the chosen identifier is unlikely to satisfy all these requirements, particularly given the limited space left in the IP header. Therefore a compromise will be necessary, which is why all the requirements are expressed with the word 'SHOULD' not 'MUST'. Appendix B discusses the pros and cons of the compromises made in various competing identification schemes against the above requirements.

On the basis of this analysis, "ECT(1) and CE codepoints" is the best compromise. Therefore this scheme is defined in detail in the following sections, while Appendix B records the rationale for this decision.

2.2. L4S Packet Identification at Run-Time

The L4S treatment is an alternative packet marking treatment [RFC4774] to the classic ECN treatment [RFC3168]. Like classic ECN, it identifies both network and host behaviour: it identifies the marking treatment that network nodes are expected to apply to L4S packets, and it identifies packets that have been sent from hosts that are expected to comply with a broad type of behaviour.

For a packet to receive L4S treatment as it is forwarded, the sender MUST set the ECN field in the IP header (v4 or v6) to the ECT(1) codepoint.

A network node that implements the L4S service MUST classify arriving ECT(1) packets for L4S treatment and it SHOULD classify arriving CE packets for L4S treatment as well. Section 2.4 describes a possible exception to this latter rule.

An L4S AQM treatment follows similar codepoint transition rules to those in RFC 3168. Specifically, the ECT(1) codepoint MUST NOT be changed to any other codepoint than CE, and CE MUST NOT be changed to any other codepoint. An ECT(1) packet is classified as ECN-capable and, if congestion increases, an L4S AQM algorithm will mark the ECN field as CE for an increasing proportion of packets, otherwise forwarding packets unchanged as ECT(1). Necessary conditions for an L4S marking treatment are defined in Section 2.5. Under persistent overload an L4S marking treatment SHOULD turn off ECN marking, using drop as a congestion signal until the overload episode has subsided, as recommended for all AQMs [RFC7567] (Section 4.2.1), which follows similar advice in RFC 3168 (Section 7).

The L4S treatment is the default for ECT(1) packets in all Diffserv Classes [RFC4774].

For backward compatibility in uncontrolled environments, a network node that implements the L4S treatment MUST also implement a classic AQM treatment. It MUST classify arriving ECT(0) and Not-ECT packets for treatment by the Classic AQM. Classic treatment means that the AQM will mark ECT(0) packets under the same conditions as it would drop Not-ECT packets [RFC3168].

2.3. Pre-Requisite Transport Layer Behaviour

2.3.1. Pre-Requisite Congestion Response

For a host to send packets with the L4S identifier (ECT(1)), it SHOULD implement a congestion control behaviour that ensures the flow rate is inversely proportional to the proportion of bytes in packets marked with the CE codepoint. This is termed a scalable congestion control, because the number of control signals (ECN marks) per round trip remains roughly constant for any flow rate. As with all transport behaviours, a detailed specification will need to be defined for each type of transport or application, including the timescale over which the proportionality is averaged, and control of burstiness. The inverse proportionality requirement above is worded as a 'SHOULD' rather than a 'MUST' to allow reasonable flexibility when defining these specifications.

Data Center TCP (DCTCP [I-D.ietf-tcpm-dctcp]) is an example of a scalable congestion control.

Each sender in a session can use a scalable congestion control independently of the congestion control used by the receiver(s) when they send data. Therefore there might be ECT(1) packets in one direction and ECT(0) in the other.

In order to coexist safely with other Internet traffic, a scalable congestion control MUST NOT identify its packets with the ECT(1) codepoint unless it complies with the following bulleted requirements. The specification of a particular scalable congestion control MUST describe in detail how it satisfies each requirement:

• A scalable congestion control MUST react to packet loss in a way that will coexist safely with a TCP Reno congestion control [RFC5681] (see Appendix A.1.3 for rationale).
• A scalable congestion control MUST react to ECN marking from a non-L4S but ECN-capable bottleneck in a way that will coexist with a TCP Reno congestion control [RFC5681] (see Appendix A.1.4 for rationale).
• A scalable congestion control MUST reduce or eliminate RTT bias over as wide a range of RTTs as possible, or at least over the typical range of RTTs that will interact in the intended deployment scenario (see Appendix A.1.5 for rationale).
• A scalable congestion control MUST remain responsive to congestion when the RTT is significantly smaller than in the current public Internet (see Appendix A.1.6 for rationale).

2.3.2. Pre-Requisite Transport Feedback

In general, a scalable congestion control needs feedback of the extent of CE marking on the forward path. Due to the history of TCP development, when ECN was added it reported no more than one CE mark per round trip. Some transport protocols derived from TCP mimic this behaviour while others report the accurate extent of TCP marking. This means that some transport protocols will need to be updated as a prerequisite for scalable congestion control. The position for a few well-known transport protocols is given below.

TCP:

Support for accurate ECN feedback (AccECN [I-D.ietf-tcpm-accurate-ecn]) by both ends is a prerequisite for scalable congestion control. Therefore, the presence of ECT(1) in the IP headers even in one direction of a TCP connection will imply that both ends support AccECN. However, the converse does not apply. So even if both ends support AccECN, either of the two ends can choose not to use a scalable congestion control, whatever the other end's choice.

SCTP:

An ECN feedback protocol such as that specified in [I-D.stewart-tsvwg-sctpecn] would be a prerequisite for scalable congestion control. That draft would update the ECN feedback protocol sketched out in Appendix A of the standards track specification of SCTP [RFC4960] by adding a field to report the number of CE marks.

RTP over UDP:

A prerequisite for scalable congestion control is for both (all) ends of one media-level hop to signal ECN support using the ecn-capable-rtp attribute [RFC6679]. Therefore, the presence of ECT(1) implies that both (all) ends of that hop support ECN. However, the converse does not apply, so each end of a media-level hop can independently choose not to use a scalable congestion control, even if both ends support ECN.

DCCP:

The ACK vector in DCCP [RFC4340] is already sufficient to report the extent of CE marking as needed by a scalable congestion control.

2.4. Exception for L4S Packet Identification by Network Nodes with Transport-Layer Awareness

To implement the L4S treatment, a network node does not need to identify transport-layer flows. Nonetheless, if an implementer is willing to identify transport-layer flows at a network node, and if the most recent ECT packet in the same flow was ECT(0), the node MAY classify CE packets for classic ECN [RFC3168] treatment. In all other cases, a network node MUST classify CE packets for L4S treatment. Examples of such other cases are: i) if no ECT packets have yet been identified in a flow; ii) if it is not desirable for a network node to identify transport-layer flows; or iii) if the most recent ECT packet in a flow was ECT(1).

If an implementer uses flow-awareness to classify CE packets, to determine whether the flow is using ECT(0) or ECT(1) it only uses the most recent ECT packet of a flow {ToDo: this advice will need to be verified experimentally}. This is because a sender might have to switch from sending ECT(1) (L4S) packets to sending ECT(0) (Classic) packets, or back again, in the middle of a transport-layer flow. Such a switch-over is likely to be very rare, but It could be necessary if the path bottleneck moves from a network node that supports L4S to one that only supports Classic ECN. A host ought to be able to detect such a change from a change in RTT variation.

2.5. The Meaning of L4S CE Relative to Drop

The likelihood that an AQM drops a Not-ECT Classic packet (p_C) MUST be roughly proportional to the square of the likelihood that it would have marked it if it had been an L4S packet (p_L). That is

• p_C ~= (p_L / k)^2

The constant of proportionality (k) does not have to be standardised for interoperability, but a value of 2 is RECOMMENDED.

[I-D.ietf-tsvwg-aqm-dualq-coupled] specifies the essential aspects of an L4S AQM, as well as recommending other aspects. It gives example implementations in appendices.

The term 'likelihood' is used above to allow for marking and dropping to be either probabilistic or deterministic. The example AQMs in [I-D.ietf-tsvwg-aqm-dualq-coupled] drop and mark probabilistically, so the drop probability is arranged to be the square of the marking probability. Nonetheless, an alternative AQM that dropped and marked deterministically would be valid, as long as the dropping frequency was proportional to the square of the marking frequency.

Note that, contrary to RFC 3168, an AQM implementing the L4S and Classic treatments does not mark an ECT(1) packet under the same conditions that it would have dropped a Not-ECT packet. However, it does mark an ECT(0) packet under the same conditions that it would have dropped a Not-ECT packet.

3. IANA Considerations

This specification contains no IANA considerations.

4. Security Considerations

Approaches to assure the integrity of signals using the new identifer are introduced in Appendix C.1.

5. Acknowledgements

Thanks to Richard Scheffenegger, John Leslie, David Täht, Jonathan Morton, Gorry Fairhurst, Michael Welzl, Mikael Abrahamsson and Andrew McGregor for the discussions that led to this specification. Ing-jyh (Inton) Tsang was a contributor to the early drafts of this document. inAppendix A listing the TCP Prague Requirements is based on text authored by Marcelo Bagnulo Braun that was originally an appendix to [I-D.ietf-tsvwg-l4s-arch]. That text was in turn based on the collective output of the attendees listed in the minutes of a 'bar BoF' on DCTCP Evolution during IETF-94 [TCPPrague].

The authors' contributions were part-funded by the European Community under its Seventh Framework Programme through the Reducing Internet Transport Latency (RITE) project (ICT-317700). The views expressed here are solely those of the authors.

6. References

6.1. Normative References

6.2. Informative References

Appendix A. The 'TCP Prague Requirements'

This appendix is informative, not normative. It gives a list of modifications to current scalable transport protocols so that they can be deployed over the public Internet and coexist safely with existing traffic. The list complements the normative requirements in Section 2.3 that a sender has to comply with before it can set the L4S identifier in packets it sends into the Internet. As well as necessary safety improvements (requirements) this appendix also includes preferable performance improvements (optimizations).

These recommendations have become know as the TCP Prague Requirements, because they were originally identified at an ad hoc meeting during IETF-94 in Prague [TCPPrague]. The wording has been generalized to apply to all scalable congestion controls, not just TCP congestion control specifically.

DCTCP [I-D.ietf-tcpm-dctcp] is currently the most widely used scalable transport protocol. In its current form, DCTCP is specified to be deployable only in controlled environments. Deploying it in the public Internet would lead to a number of issues, both from the safety and the performance perspective. The modifications and additional mechanisms listed in this section will be necessary for its deployment over the global Internet. Where an example is needed, DCTCP is used as a base, but it is likely that most of these requirements equally apply to other scalable transport protocols.

A.1. Requirements for Scalable Transport Protocols

A.1.1. Use of L4S Packet Identifier

Description: A scalable congestion control needs to distinguish the packets it sends from those sent by classic congestion controls.

Motivation: It needs to be possible for a network node to classify L4S packets without flow state into a queue that applies an L4S ECN marking behaviour and isolates L4S packets from the queuing delay of classic packets.

A.1.2. Accurate ECN Feedback

Description: A scalable transport protocol needs to provide timely, accurate feedback about the extent of ECN marking experienced by all packets.

Motivation: Classic congestion controls only need feedback about the existence of a congestion episode within a round trip, not precisely how many packets were marked with ECN or dropped. Therefore, in 2001, when ECN feedback was added to TCP [RFC3168], it could not inform the sender of more than one ECN mark per RTT. Since then, requirements for more accurate ECN feedback in TCP have been defined in [RFC7560] and [I-D.ietf-tcpm-accurate-ecn] specifies an experimental change to the TCP wire protocol to satisfy these requirements. Most other transport protocols already satisfy this requirement.

A.1.3. Fall back to Reno/Cubic congestion control on packet loss

Description: A scalable congestion control needs to react to packet loss in a way that will coexist safely with a TCP Reno congestion control [RFC5681].

Motivation: Part of the safety conditions for deploying a scalable congestion control on the public Internet is to make sure that it behaves properly when it builds a queue at a network bottleneck that has not been upgraded to support L4S. Packet loss can have many causes, but it usually has to be conservatively assumed that it is a sign of congestion. Therefore, on detecting packet loss, a scalable congestion control will need to fall back to classic congestion control behaviour. If it does not comply with this requirement it could starve classic traffic.

A scalable congestion control can be used for different types of transport, e.g. for real-time media or for reliable bulk transport like TCP. Therefore, the particular classic congestion control behaviour to fall back on will need to be part of the congestion control specification of the relevant transport. In the particular case of DCTCP, the current DCTCP specification states that "It is RECOMMENDED that an implementation deal with loss episodes in the same way as conventional TCP." For safe deployment of a scalable transport in the public Internet, the above requirement would need to be defined as a "MUST".

Packet loss might (rarely) occur in the case that the bottleneck is L4S capable. In this case, the sender may receive a high number of packets marked with the CE bit set and also experience a loss. Current DCTCP implementations react differently to this situation. At least one implementation reacts only to the drop signal (e.g. by halving the CWND) and at least another DCTCP implementation reacts to both signals (e.g. by halving the CWND due to the drop and also further reducing the CWND based on the proportion of marked packet). We believe that further experimentation is needed to understand what is the best behaviour for the public Internet, which may or not be one of these existing approaches.

A.1.4. Fall back to Reno/Cubic congestion control on classic ECN bottlenecks

Description: A scalable congestion control needs to react to ECN marking from a non-L4S but ECN-capable bottleneck in a way that will coexist with a TCP Reno congestion control [RFC5681].

Motivation: Similarly to the requirement in Appendix A.1.3, this requirement is a safety condition to ensure a scalable congestion control behaves properly when it builds a queue at a network bottleneck that has not been upgraded to support L4S. On detecting classic ECN marking (see below), a scalable congestion control will need to fall back to classic congestion control behaviour. If it does not comply with this requirement it could starve classic traffic.

It would take time for endpoints to distinguish classic and L4S ECN marking. An increase in queuing delay or in delay variation would be a tell-tale sign, but it is not yet clear where a line would be drawn between the two behaviours. It might be possible to cache what was learned about the path to help subsequent attempts to detect the type of marking.

A.1.5. Reduce RTT dependence

Description: A scalable congestion control needs to reduce or eliminate RTT bias over as wide a range of RTTs as possible, or at least over the typical range of RTTs that will interact in the intended deployment scenario.

Motivation: Classic TCP's throughput is known to be inversely proportional to RTT, so one would expect flows over very low RTT paths to nearly starve flows over larger RTTs. However, Classic TCP has never allowed a very low RTT path to exist because it induces a large queue. For instance, consider two paths with base RTT 1ms and 100ms. If Classic TCP induces a 100ms queue, it turns these RTTs into 101ms and 200ms leading to a throughput ratio of about 2:1. Whereas if a Scalable TCP induces only a 1ms queue, the ratio is 2:101, leading to a throughput ratio of about 50:1.

Therefore, with very small queues, long RTT flows will essentially starve, unless scalable congestion controls comply with this requirement.

A.1.6. Scaling down the congestion window

Description: A scalable congestion control needs to remain responsive to congestion when RTTs are significantly smaller than in the current public Internet.

Motivation: As currently specified, the minimum required congestion window of TCP (and its derivatives) is set to 2 maximum segment sizes (MSS) (see equation (4) in [RFC5681]). Once the congestion window reaches this minimum, all current TCP algorithms become unresponsive to congestion signals. No matter how much drop or ECN marking, the congestion window no longer reduces. Instead, TCP forces the queue to grow, overriding any AQM and increasing queuing delay.

L4S mechanisms significantly reduce queueing delay so, over the same path, the RTT becomes lower. Then this problem becomes surprisingly common [TCP-sub-mss-w]. This is because, for the same link capacity, smaller RTT implies a smaller window. For instance, consider a residential setting with an upstream broadband Internet access of 8 Mb/s, assuming a max segment size of 1500 B. Two upstream flows will each have the minimum window of 2 MSS if the RTT is 6ms or less, which is quite common when accessing a nearby data centre. So, any more than two such parallel TCP flows will become unresponsive and increase queuing delay.

Unless scalable congestion controls are required to comply with this requirement from the start, they will frequently become unresponsive, negating the low latency benefit of L4S, for themselves and for others. One possible sub-MSS window mechanism is described in [TCP-sub-mss-w], and other approaches are likely to be feasible.

A.2. Scalable Transport Protocol Optimizations

A.2.1. Setting ECT in TCP Control Packets and Retransmissions

Description: To improve performance, scalable transport protocols ought to enable ECN at the IP layer in TCP control packets (SYN, SYN-ACK, pure ACKs, etc.) and in retransmitted packets. The same is true for derivatives of TCP, e.g. SCTP.

Motivation: RFC 3168 prohibits the use of ECN on these types of TCP packet, based on a number of arguments. This means these packets are not protected from congestion loss by ECN, which considerably harms performance, particularly for short flows. [I-D.bagnulo-tcpm-generalized-ecn] counters each argument in RFC 3168 in turn, showing it was over-cautious. Instead it proposes experimental use of ECN on all types of TCP packet.

A.2.2. Faster than Additive Increase

Description: It would improve performance if scalable congestion controls did not limit their congestion window increase to the traditional additive increase of 1 MSS per round trip [RFC5681] during congestion avoidance. The same is true for derivatives of TCP congestion control.

Motivation: As currently defined, DCTCP uses the traditional TCP Reno additive increase in congestion avoidance phase. When the available capacity suddenly increases (e.g. when another flow finishes, or if radio capacity increases) it can take very many round trips to take advantage of the new capacity. In the steady state, DCTCP induces about 2 ECN marks per round trip, so it should be possible to quickly detect when these signals have disappeared and seek available capacity more rapidly. It will of course be necessary to minimize the impact on other flows (classic and scalable).

TCP Cubic was designed to solve this problem, but as flow rates have continued to increase, the delay accelerating into available capacity has become prohibitive. For instance, with RTT=20 ms, to increase flow rate from 100Mb/s to 200Mb/s Cubic takes between 50 and 100 round trips. Every 8x increase in flow rate leads to 2x more acceleration delay.

A.2.3. Faster Convergence at Flow Start

Description: Particularly when a flow starts, scalable congestion controls need to converge (reach their steady-state share of the capacity) at least as fast as classic TCP and preferably faster. This affects the flow start behaviour of all transports that are based on TCP slow start.

Motivation: As an example, a new DCTCP flow takes longer than classic TCP to obtain its share of the capacity of the bottleneck when there are already ongoing flows using the bottleneck capacity (about a factor of 1.5 and 2 longer according to [Alizadeh-stability]). This is detrimental in general, but particularly harms the performance of short flows relative to classic TCP.

Appendix B. Alternative Identifiers

This appendix is informative, not normative. It records the pros and cons of various alternative ways to identify L4S packets to record the rationale for the choice of ECT(1) (Appendix B.1) as the L4S identifier. At the end, Appendix B.6 summarises the distinguishing features of the leading alternatives. It is intended to supplement, not replace the detailed text.

The leading solutions all use the ECN field, sometimes in combination with the Diffserv field. Both the ECN and Diffserv fields have the additional advantage that they are no different in either IPv4 or IPv6. A couple of alternatives that use other fields are mentioned at the end, but it is quickly explained why they are not serious contenders.

B.1. ECT(1) and CE codepoints

Definition:

• Packets with ECT(1) and conditionally packets with CE would signify L4S semantics as an alternative to the semantics of classic ECN [RFC3168], specifically:
  • The ECT(1) codepoint would signify that the packet was sent by an L4S-capable sender;
  • Given shortage of codepoints, both L4S and classic ECN sides of an AQM would have to use the same CE codepoint to indicate that a packet had experienced congestion. If a packet that had already been marked CE in an upstream buffer arrived at a subsequent AQM, this AQM would then have to guess whether to classify CE packets as L4S or classic ECN. Choosing the L4S treatment would be a safer choice, because then a few classic packets might arrive early, rather than a few L4S packets arriving late;
  • Additional information might be available if the classifier were transport-aware. Then it could classify a CE packet for classic ECN treatment if the most recent ECT packet in the same flow had been marked ECT(0). However, the L4S service ought not to need tranport-layer awareness;

Cons:

Consumes the last ECN codepoint:

The L4S service is intended to supersede the service provided by classic ECN, therefore using ECT(1) to identify L4S packets could ultimately mean that the ECT(0) codepoint was 'wasted' purely to distinguish one form of ECN from its successor;

ECN hard in some lower layers:

It is not always possible to support ECN in an AQM acting in a buffer below the IP layer [I-D.ietf-tsvwg-ecn-encap-guidelines]. In such cases, the L4S service would have to drop rather than mark frames even though they might contain an ECN-capable packet. However, such cases would be unusual.

Risk of reordering classic CE packets:

Having to classify all CE packets as L4S risks some classic CE packets arriving early, which is a form of reordering. Reordering can cause the TCP sender to retransmit spuriously. However, one or two packets delivered early does not cause any spurious retransmissions because the subsequent packets continue to move the cumulative acknowledgement boundary forwards. Anyway, the risk of reordering would be low, because: i) it is quite unusual to experience more than one bottleneck queue on a path; ii) even then, reordering would only occur if there was simultaneous mixing of classic and L4S traffic, which would be more unlikely in an access link, which is where most bottlenecks are located; iii) even then, spurious retransmissions would only occur if a contiguous sequence of three or more classic CE packets from one bottleneck arrived at the next, which should in itself happen very rarely with a good AQM. The risk would be completely eliminated in AQMs that were transport-aware (but they should not need to be);

Non-L4S service for control packets:

The classic ECN RFCs [RFC3168] and [RFC5562] require a sender to clear the ECN field to Not-ECT for retransmissions and certain control packets specifically pure ACKs, window probes and SYNs. When L4S packets are classified by the ECN field alone, these control packets would not be classified into an L4S queue, and could therefore be delayed relative to the other packets in the flow. This would not cause re-ordering (because retransmissions are already out of order, and the control packets carry no data). However, it would make critical control packets more vulnerable to loss and delay. To address this problem, [I-D.bagnulo-tswg-generalized-ecn] proposes an experiment in which all TCP control packets and retransmissions are ECN-capable.

Pros:

Should work e2e:

The ECN field generally works end-to-end across the Internet. Unlike the DSCP, the setting of the ECN field is at least forwarded unchanged by networks that do not support ECN, and networks rarely clear it to zero;

Should work in tunnels:

Unlike Diffserv, ECN is defined to always work across tunnels. However, tunnels do not always implement ECN processing as they should do, particularly because IPsec tunnels were defined differently for a few years.

Could migrate to one codepoint:

If all classic ECN senders eventually evolve to use the L4S service, the ECT(0) codepoint could be reused for some future purpose, but only once use of ECT(0) packets had reduced to zero, or near-zero, which might never happen.

B.2. ECN Plus a Diffserv Codepoint (DSCP)

Definition:

• For packets with a defined DSCP, all codepoints of the ECN field (except Not-ECT) would signify alternative L4S semantics to those for classic ECN [RFC3168], specifically:
  • The L4S DSCP would signifiy that the packet came from an L4S-capable sender;
  • ECT(0) and ECT(1) would both signify that the packet was travelling between transport endpoints that were both ECN-capable;
  • CE would signify that the packet had been marked by an AQM implementing the L4S service.

Use of a DSCP is the only approach for alternative ECN semantics given as an example in [RFC4774]. However, it was perhaps considered more for controlled environments than new end-to-end services;

Cons:

Consumes DSCP pairs:

A DSCP is obviously not orthogonal to Diffserv. Therefore, wherever the L4S service is applied to multiple Diffserv scheduling behaviours, it would be necessary to replace each DSCP with a pair of DSCPs.

Uses critical lower-layer header space:

The resulting increased number of DSCPs might be hard to support for some lower layer technologies, e.g. 802.1p and MPLS both offer only 3-bits for a maximum of 8 traffic class identifiers. Although L4S should reduce and possibly remove the need for some DSCPs intended for differentiated queuing delay, it will not remove the need for Diffserv entirely, because Diffserv is also used to allocate bandwidth, e.g. by prioritising some classes of traffic over others when traffic exceeds available capacity.

Not end-to-end (host-network):

Very few networks honour a DSCP set by a host. Typically a network will zero (bleach) the Diffserv field from all hosts. Sometimes networks will attempt to identify applications by some form of packet inspection and, based on network policy, they will set the DSCP considered appropriate for the identified application. Network-based application identification might use some combination of protocol ID, port numbers(s), application layer protocol headers, IP address(es), VLAN ID(s) and even packet timing.

Not end-to-end (network-network):

Very few networks honour a DSCP received from a neighbouring network. Typically a network will zero (bleach) the Diffserv field from all neighbouring networks at an interconnection point. Sometimes bilateral arrangements are made between networks, such that the receiving network remarks some DSCPs to those it uses for roughly equivalent services. The likelihood that a DSCP will be bleached or ignored depends on the type of DSCP:

Local-use DSCP:

These tend to be used to implement application-specific network policies, but a bilateral arrangement to remark certain DSCPs is often applied to DSCPs in the local-use range simply because it is easier not to change all of a network's internal configurations when a new arrangement is made with a neighbour;

Global-use DSCP:

These do not tend to be honoured across network interconnections more than local-use DSCPs. However, if two networks decide to honour certain of each other's DSCPs, the reconfiguration is a little easier if both of their globally recognised services are already represented by the relevant global-use DSCPs.

Note that today a global-use DSCP gives little more assurance of end-to-end service than a local-use DSCP. In future the global-use range might give more assurance of end-to-end service than local-use, but it is unlikely that either assurance will be high, particularly given the hosts are included in the end-to-end path.

Not all tunnels:

Diffserv codepoints are often not propagated to the outer header when a packet is encapsulated by a tunnel header. DSCPs are propagated to the outer of uniform mode tunnels, but not pipe mode [RFC2983], and pipe mode is fairly common.

ECN hard in some lower layers::

Because this approach uses both the Diffserv and ECN fields, an AQM wil only work at a lower layer if both can be supported. If individual network operators wished to deploy an AQM at a lower layer, they would usually propagate an IP Diffserv codepoint to the lower layer, using for example IEEE 802.1p. However, the ECN capability is harder to propagate down to lower layers because few lower layers support it.

Pros:

Could migrate to e2e:

If all usage of classic ECN migrates to usage of L4S, the DSCP would become redundant, and the ECN capability alone could eventually identify L4S packets without the interconnection problems of Diffserv detailed above, and without having permanently consumed more than one codepoint in the IP header. Although the DSCP does not generally function as an end-to-end identifier (see above), it could be used initially by individual ISPs to introduce the L4S service for their own locally generated traffic;

B.3. ECN capability alone

Definition:

• This approach uses ECN capability alone as the L4S identifier. It is only feasible if classic ECN is not widely deployed. The specific definition of codepoints would be:
  • Any ECN codepoint other than Not-ECT would signify an L4S-capable sender;
  • ECN codepoints would not be used for classic [RFC3168] ECN, and the classic network service would only be used for Not-ECT packets.

  This approach would only be feasible if

  1. it was generally agreed that there was little chance of any classic [RFC3168] ECN deployment in any network nodes;
  2. it was generally agreed that there was little chance of any client devices being deployed with classic [RFC3168] TCP-ECN on by default (note that classic TCP-ECN is already on-by-default on many servers);
  3. for TCP connections, developers of client OSs would all have to agree not to encourage further deployment of classic ECN. Specifically, at the start of a TCP connection classic ECN could be disabled during negotation of the ECN capability:
     • an L4S-capable host would have to disable ECN if the corresponding host did not support accurate ECN feedback [RFC7560], which is a prerequisite for the L4S service;
     • developers of operating systems for user devices would only enable ECN by default for TCP once the stack implemented L4S and accurate ECN feedback [RFC7560] including requesting accurate ECN feedback by default.

Cons:

Near-infeasible deployment constraints:

The constraints for deployment above represent a highly unlikely, but not completely impossible, set of circumstances. If, despite the above measures, a pair of hosts did negotiate to use classic ECN, their packets would be classified into the same queue as L4S traffic, and if they had to compete with a long-running L4S flow they would get a very small capacity share;

ECN hard in some lower layers:

See the same issue with "ECT(1) and CE codepoints" (Appendix B.1);

Non-L4S service for control packets:

See the same issue with "ECT(1) and CE codepoints" (Appendix B.1).

Pros:

Consumes no additional codepoints:

The ECT(1) codepoint and all spare Diffserv codepoints would remain available for future use;

Should work e2e:

As with "ECT(1) and CE codepoints" (Appendix B.1);

Should work in tunnels:

As with "ECT(1) and CE codepoints" (Appendix B.1).

B.4. Protocol ID

It has been suggested that a new ID in the IPv4 Protocol field or the IPv6 Next Header field could identify L4S packets. However this approach is ruled out by numerous problems:

• A new protocol ID would need to be paired with the old one for each transport (TCP, SCTP, UDP, etc.);
• In IPv6, there can be a sequence of Next Header fields, and it would not be obvious which one would be expected to identify a network service like L4S;
• A new protocol ID would rarely provide an end-to-end service, because It is well-known that new protocol IDs are often blocked by numerous types of middlebox;
• The approach is not a solution for AQMs below the IP layer;

B.5. Source or destination addressing

Locally, a network operator could arrange for L4S service to be applied based on source or destination addressing, e.g. packets from its own data centre and/or CDN hosts, packets to its business customers, etc. It could use addressing at any layer, e.g. IP addresses, MAC addresses, VLAN IDs, etc. Although addressing might be a useful tactical approach for a single ISP, it would not be a feasible approach to identify an end-to-end service like L4S. Even for a single ISP, it would require packet classifiers in buffers to be dependent on changing topology and address allocation decisions elsewhere in the network. Therefore this approach is not a feasible solution.

B.6. Summary: Merits of Alternative Identifiers

Table 1 provides a very high level summary of the pros and cons detailed against the schemes described respectively in Appendix B.2, Appendix B.3 and Appendix B.1, for six issues that set them apart.