QUIC J. Iyengar, Ed.
Internet-Draft I. Swett, Ed.
Intended status: Standards Track Google
Expires: November 23, 2017 May 22, 2017

QUIC Loss Detection and Congestion Control


This document describes loss detection and congestion control mechanisms for QUIC.

Note to Readers

Discussion of this draft takes place on the QUIC working group mailing list (quic@ietf.org), which is archived at https://mailarchive.ietf.org/arch/search/?email_list=quic.

Working Group information can be found at https://github.com/quicwg; source code and issues list for this draft can be found at https://github.com/quicwg/base-drafts/labels/recovery.

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 23, 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

QUIC is a new multiplexed and secure transport atop UDP. QUIC builds on decades of transport and security experience, and implements mechanisms that make it attractive as a modern general-purpose transport. The QUIC protocol is described in [QUIC-TRANSPORT].

QUIC implements the spirit of known TCP loss recovery mechanisms, described in RFCs, various Internet-drafts, and also those prevalent in the Linux TCP implementation. This document describes QUIC congestion control and loss recovery, and where applicable, attributes the TCP equivalent in RFCs, Internet-drafts, academic papers, and/or TCP implementations.

1.1. Notational Conventions

The words “MUST”, “MUST NOT”, “SHOULD”, and “MAY” are used in this document. It’s not shouting; when they are capitalized, they have the special meaning defined in [RFC2119].

2. Design of the QUIC Transmission Machinery

All transmissions in QUIC are sent with a packet-level header, which includes a packet sequence number (referred to below as a packet number). These packet numbers never repeat in the lifetime of a connection, and are monotonically increasing, which makes duplicate detection trivial. This fundamental design decision obviates the need for disambiguating between transmissions and retransmissions and eliminates significant complexity from QUIC’s interpretation of TCP loss detection mechanisms.

Every packet may contain several frames. We outline the frames that are important to the loss detection and congestion control machinery below.

2.1. Relevant Differences Between QUIC and TCP

Readers familiar with TCP’s loss detection and congestion control will find algorithms here that parallel well-known TCP ones. Protocol differences between QUIC and TCP however contribute to algorithmic differences. We briefly describe these protocol differences below.

2.1.1. Monotonically Increasing Packet Numbers

TCP conflates transmission sequence number at the sender with delivery sequence number at the receiver, which results in retransmissions of the same data carrying the same sequence number, and consequently to problems caused by “retransmission ambiguity”. QUIC separates the two: QUIC uses a packet sequence number (referred to as the “packet number”) for transmissions, and any data that is to be delivered to the receiving application(s) is sent in one or more streams, with stream offsets encoded within STREAM frames inside of packets that determine delivery order.

QUIC’s packet number is strictly increasing, and directly encodes transmission order. A higher QUIC packet number signifies that the packet was sent later, and a lower QUIC packet number signifies that the packet was sent earlier. When a packet containing frames is deemed lost, QUIC rebundles necessary frames in a new packet with a new packet number, removing ambiguity about which packet is acknowledged when an ACK is received. Consequently, more accurate RTT measurements can be made, spurious retransmissions are trivially detected, and mechanisms such as Fast Retransmit can be applied universally, based only on packet number.

This design point significantly simplifies loss detection mechanisms for QUIC. Most TCP mechanisms implicitly attempt to infer transmission ordering based on TCP sequence numbers - a non-trivial task, especially when TCP timestamps are not available.

2.1.2. No Reneging

QUIC ACKs contain information that is equivalent to TCP SACK, but QUIC does not allow any acked packet to be reneged, greatly simplifying implementations on both sides and reducing memory pressure on the sender.

2.1.3. More ACK Ranges

QUIC supports up to 256 ACK ranges, opposed to TCP’s 3 SACK ranges. In high loss environments, this speeds recovery.

2.1.4. Explicit Correction For Delayed Acks

QUIC ACKs explicitly encode the delay incurred at the receiver between when a packet is received and when the corresponding ACK is sent. This allows the receiver of the ACK to adjust for receiver delays, specifically the delayed ack timer, when estimating the path RTT. This mechanism also allows a receiver to measure and report the delay from when a packet was received by the OS kernel, which is useful in receivers which may incur delays such as context-switch latency before a userspace QUIC receiver processes a received packet.

3. Loss Detection

3.1. Overview

QUIC uses a combination of ack information and alarms to detect lost packets. An unacknowledged QUIC packet is marked as lost in one of the following ways:

3.2. Algorithm Details

3.2.1. Constants of interest

Constants used in loss recovery are based on a combination of RFCs, papers, and common practice. Some may need to be changed or negotiated in order to better suit a variety of environments.

kMaxTLPs (default 2):
Maximum number of tail loss probes before an RTO fires.
kReorderingThreshold (default 3):
Maximum reordering in packet number space before FACK style loss detection considers a packet lost.
kTimeReorderingFraction (default 1/8):
Maximum reordering in time space before time based loss detection considers a packet lost. In fraction of an RTT.
kMinTLPTimeout (default 10ms):
Minimum time in the future a tail loss probe alarm may be set for.
kMinRTOTimeout (default 200ms):
Minimum time in the future an RTO alarm may be set for.
kDelayedAckTimeout (default 25ms):
The length of the peer’s delayed ack timer.
kDefaultInitialRtt (default 100ms):
The default RTT used before an RTT sample is taken.

3.2.2. Variables of interest

Variables required to implement the congestion control mechanisms are described in this section.

Multi-modal alarm used for loss detection.
The number of times the handshake packets have been retransmitted without receiving an ack.
The number of times a tail loss probe has been sent without receiving an ack.
The number of times an rto has been sent without receiving an ack.
The last packet number sent prior to the first retransmission timeout.
The smoothed RTT of the connection, computed as described in [RFC6298]
The RTT variance, computed as described in [RFC6298]
The largest delta between the largest acked retransmittable packet and a packet containing retransmittable frames before it’s declared lost.
The reordering window as a fraction of max(smoothed_rtt, latest_rtt).
The time at which the next packet will be considered lost based on early transmit or exceeding the reordering window in time.
An association of packet numbers to information about them, including a number field indicating the packet number, a time field indicating the time a packet was sent, and a bytes field indicating the packet’s size. sent_packets is ordered by packet number, and packets remain in sent_packets until acknowledged or lost.

3.2.3. Initialization

At the beginning of the connection, initialize the loss detection variables as follows:

   handshake_count = 0
   tlp_count = 0
   rto_count = 0
   if (UsingTimeLossDetection())
     reordering_threshold = infinite
     time_reordering_fraction = kTimeReorderingFraction
     reordering_threshold = kReorderingThreshold
     time_reordering_fraction = infinite
   loss_time = 0
   smoothed_rtt = 0
   rttvar = 0
   largest_sent_before_rto = 0

3.2.4. On Sending a Packet

After any packet is sent, be it a new transmission or a rebundled transmission, the following OnPacketSent function is called. The parameters to OnPacketSent are as follows:

Pseudocode for OnPacketSent follows:

 OnPacketSent(packet_number, is_retransmittable, sent_bytes):
   sent_packets[packet_number].packet_number = packet_number
   sent_packets[packet_number].time = now
   if is_retransmittable:
     sent_packets[packet_number].bytes = sent_bytes

3.2.5. On Ack Receipt

When an ack is received, it may acknowledge 0 or more packets.

The sender MUST abort the connection if it receives an ACK for a packet it never sent, see [QUIC-TRANSPORT].

Pseudocode for OnAckReceived and UpdateRtt follow:

     // If the largest acked is newly acked, update the RTT.
     if (sent_packets[ack.largest_acked]):
       rtt_sample = now - sent_packets[ack.largest_acked].time
       if (rtt_sample > ack.ack_delay):
         rtt_sample -= ack.delay
     // The sender may skip packets for detecting optimistic ACKs
     if (packets acked that the sender skipped):
     // Find all newly acked packets.
     for acked_packet in DetermineNewlyAckedPackets():


     // Based on {{RFC6298}}.
     if (smoothed_rtt == 0):
       smoothed_rtt = rtt_sample
       rttvar = rtt_sample / 2
       rttvar = 3/4 * rttvar + 1/4 * (smoothed_rtt - rtt_sample)
       smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * rtt_sample

3.2.6. On Packet Acknowledgment

When a packet is acked for the first time, the following OnPacketAcked function is called. Note that a single ACK frame may newly acknowledge several packets. OnPacketAcked must be called once for each of these newly acked packets.

OnPacketAcked takes one parameter, acked_packet, which is the packet number of the newly acked packet, and returns a list of packet numbers that are detected as lost.

If this is the first acknowledgement following RTO, check if the smallest newly acknowledged packet is one sent by the RTO, and if so, inform congestion control of a verified RTO, similar to F-RTO [RFC5682]

Pseudocode for OnPacketAcked follows:

     // If a packet sent prior to RTO was acked, then the RTO
     // was spurious.  Otherwise, inform congestion control.
     if (rto_count > 0 &&
         acked_packet_number > largest_sent_before_rto)
     handshake_count = 0
     tlp_count = 0
     rto_count = 0

3.2.7. Setting the Loss Detection Alarm

QUIC loss detection uses a single alarm for all timer-based loss detection. The duration of the alarm is based on the alarm’s mode, which is set in the packet and timer events further below. The function SetLossDetectionAlarm defined below shows how the single timer is set based on the alarm mode. Handshake Packets

The initial flight has no prior RTT sample. A client SHOULD remember the previous RTT it observed when resumption is attempted and use that for an initial RTT value. If no previous RTT is available, the initial RTT defaults to 200ms.

Endpoints MUST retransmit handshake frames if not acknowledged within a time limit. This time limit will start as the largest of twice the rtt value and MinTLPTimeout. Each consecutive handshake retransmission doubles the time limit, until an acknowledgement is received.

Handshake frames may be cancelled by handshake state transitions. In particular, all non-protected frames SHOULD be no longer be transmitted once packet protection is available.

When stateless rejects are in use, the connection is considered immediately closed once a reject is sent, so no timer is set to retransmit the reject.

Version negotiation packets are always stateless, and MUST be sent once per per handshake packet that uses an unsupported QUIC version, and MAY be sent in response to 0RTT packets. Tail Loss Probe and Retransmission Timeout

Tail loss probes [LOSS-PROBE] and retransmission timeouts [RFC6298] are an alarm based mechanism to recover from cases when there are outstanding retransmittable packets, but an acknowledgement has not been received in a timely manner. Early Retransmit

Early retransmit [RFC5827] is implemented with a 1/4 RTT timer. It is part of QUIC’s time based loss detection, but is always enabled, even when only packet reordering loss detection is enabled. Pseudocode

Pseudocode for SetLossDetectionAlarm follows:

    if (retransmittable packets are not outstanding):

    if (handshake packets are outstanding):
      // Handshake retransmission alarm.
      if (smoothed_rtt == 0):
        alarm_duration = 2 * kDefaultInitialRtt
        alarm_duration = 2 * smoothed_rtt
      alarm_duration = max(alarm_duration, kMinTLPTimeout)
      alarm_duration = alarm_duration * (2 ^ handshake_count)
    else if (loss_time != 0):
      // Early retransmit timer or time loss detection.
      alarm_duration = loss_time - now
    else if (tlp_count < kMaxTLPs):
      // Tail Loss Probe
      if (retransmittable_packets_outstanding = 1):
        alarm_duration = 1.5 * smoothed_rtt + kDelayedAckTimeout
        alarm_duration = kMinTLPTimeout
      alarm_duration = max(alarm_duration, 2 * smoothed_rtt)
      // RTO alarm
      alarm_duration = smoothed_rtt + 4 * rttvar
      alarm_duration = max(alarm_duration, kMinRTOTimeout)
      alarm_duration = alarm_duration * (2 ^ rto_count)

    loss_detection_alarm.set(now + alarm_duration)

3.2.8. On Alarm Firing

QUIC uses one loss recovery alarm, which when set, can be in one of several modes. When the alarm fires, the mode determines the action to be performed.

Pseudocode for OnLossDetectionAlarm follows:

     if (handshake packets are outstanding):
       // Handshake retransmission alarm.
     else if (loss_time != 0):
       // Early retransmit or Time Loss Detection
     else if (tlp_count < kMaxTLPs):
       // Tail Loss Probe.
       // RTO.
       if (rto_count == 0)
         largest_sent_before_rto = largest_sent_packet


3.2.9. Detecting Lost Packets

Packets in QUIC are only considered lost once a larger packet number is acknowledged. DetectLostPackets is called every time an ack is received. If the loss detection alarm fires and the loss_time is set, the previous largest acked packet is supplied. Handshake Packets

The receiver MUST ignore unprotected packets that ack protected packets. The receiver MUST trust protected acks for unprotected packets, however. Aside from this, loss detection for handshake packets when an ack is processed is identical to other packets. Pseudocode

DetectLostPackets takes one parameter, acked, which is the largest acked packet.

Pseudocode for DetectLostPackets follows:

     loss_time = 0
     lost_packets = {}
     delay_until_lost = infinite
     if (time_reordering_fraction != infinite):
       delay_until_lost =
         (1 + time_reordering_fraction) * max(latest_rtt, smoothed_rtt)
     else if (largest_acked.packet_number == largest_sent_packet):
       // Early retransmit alarm.
       delay_until_lost = 9/8 * max(latest_rtt, smoothed_rtt)
     foreach (unacked < largest_acked.packet_number):
       time_since_sent = now() - unacked.time_sent
       packet_delta = largest_acked.packet_number - unacked.packet_number
       if (time_since_sent > delay_until_lost):
       else if (packet_delta > reordering_threshold)
       else if (loss_time == 0 && delay_until_lost != infinite):
         loss_time = now() + delay_until_lost - time_since_sent

     // Inform the congestion controller of lost packets and
     // lets it decide whether to retransmit immediately.
     if (!lost_packets.empty())
       foreach (packet in lost_packets)

3.3. Discussion

The majority of constants were derived from best common practices among widely deployed TCP implementations on the internet. Exceptions follow.

A shorter delayed ack time of 25ms was chosen because longer delayed acks can delay loss recovery and for the small number of connections where less than packet per 25ms is delivered, acking every packet is beneficial to congestion control and loss recovery.

The default initial RTT of 100ms was chosen because it is slightly higher than both the median and mean min_rtt typically observed on the public internet.

4. Congestion Control

QUIC’s congestion control is based on TCP NewReno[RFC6582] congestion control to determine the congestion window and pacing rate.

4.1. Slow Start

QUIC begins every connection in slow start and exits slow start upon loss. While in slow start, QUIC increases the congestion window by the number of acknowledged bytes when each ack is processed.

4.2. Recovery

Recovery is a period of time beginning with detection of a lost packet. It ends when all packets outstanding at the time recovery began have been acknowledged or lost. During recovery, the congestion window is not increased or decreased.

4.3. Constants of interest

Constants used in congestion control are based on a combination of RFCs, papers, and common practice. Some may need to be changed or negotiated in order to better suit a variety of environments.

kDefaultMss (default 1460 bytes):
The default max packet size used for calculating default and minimum congestion windows.
kInitialWindow (default 10 * kDefaultMss):
Default limit on the amount of outstanding data in bytes.
kMinimumWindow (default 2 * kDefaultMss):
Default minimum congestion window.
kLossReductionFactor (default 0.5):
Reduction in congestion window when a new loss event is detected.

4.4. Variables of interest

Variables required to implement the congestion control mechanisms are described in this section.

The sum of the size in bytes of all sent packets that contain at least one retransmittable frame, and have not been acked or declared lost.
Maximum number of bytes in flight that may be sent.
The packet number after which QUIC will no longer be in recovery.
Slow start threshold in bytes. When the congestion window is below ssthresh, it grows by the number of bytes acknowledged for each ack.

4.5. Initialization

At the beginning of the connection, initialize the loss detection variables as follows:

   congestion_window = kInitialWindow
   bytes_in_flight = 0
   end_of_recovery = 0
   ssthresh = infinite

4.6. On Packet Acknowledgement

Invoked at the same time loss detection’s OnPacketAcked is called and supplied with the acked_packet from sent_packets.

Pseudocode for OnPacketAcked follows:

     if (acked_packet.packet_number < end_of_recovery):
     if (congestion_window < ssthresh):
       congestion_window += acket_packets.bytes
       congestion_window +=
           acked_packets.bytes / congestion_window

4.7. On Packets Lost

Invoked by loss detection from DetectLostPackets when new packets are detected lost.

     largest_lost_packet = lost_packets.last()
     // Start a new recovery epoch if the lost packet is larger
     // than the end of the previous recovery epoch.
     if (end_of_recovery < largest_lost_packet.packet_number):
       end_of_recovery = largest_sent_packet
       congestion_window *= kLossReductionFactor
       congestion_window = max(congestion_window, kMinimumWindow)
       ssthresh = congestion_window

4.8. On Retransmission Timeout Verified

QUIC decreases the congestion window to the minimum value once the retransmission timeout has been confirmed to not be spurious when the first post-RTO acknowledgement is processed.

     congestion_window = kMinimumWindow

4.9. Pacing Packets

QUIC sends a packet if there is available congestion window and sending the packet does not exceed the pacing rate.

TimeToSend returns infinite if the congestion controller is congestion window limited, a time in the past if the packet can be sent immediately, and a time in the future if sending is pacing limited.

     if (bytes_in_flight + packet_size > congestion_window)
       return infinite
     return time_of_last_sent_packet +
         (packet_size * smoothed_rtt) / congestion_window

5. IANA Considerations

This document has no IANA actions. Yet.

6. References

6.1. Normative References

[QUIC-TRANSPORT] Iyengar, J. and M. Thomson, "QUIC: A UDP-Based Multiplexed and Secure Transport", Internet-Draft draft-ietf-quic-transport, May 2017.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997.

6.2. Informative References

[LOSS-PROBE] Dukkipati, N., Cardwell, N., Cheng, Y. and M. Mathis, "Tail Loss Probe (TLP): An Algorithm for Fast Recovery of Tail Losses", Internet-Draft draft-dukkipati-tcpm-tcp-loss-probe-01, February 2013.
[RFC5682] Sarolahti, P., Kojo, M., Yamamoto, K. and M. Hata, "Forward RTO-Recovery (F-RTO): An Algorithm for Detecting Spurious Retransmission Timeouts with TCP", RFC 5682, DOI 10.17487/RFC5682, September 2009.
[RFC5827] Allman, M., Avrachenkov, K., Ayesta, U., Blanton, J. and P. Hurtig, "Early Retransmit for TCP and Stream Control Transmission Protocol (SCTP)", RFC 5827, DOI 10.17487/RFC5827, May 2010.
[RFC6298] Paxson, V., Allman, M., Chu, J. and M. Sargent, "Computing TCP's Retransmission Timer", RFC 6298, DOI 10.17487/RFC6298, June 2011.
[RFC6582] Henderson, T., Floyd, S., Gurtov, A. and Y. Nishida, "The NewReno Modification to TCP's Fast Recovery Algorithm", RFC 6582, DOI 10.17487/RFC6582, April 2012.

Appendix A. Acknowledgments

Appendix B. Change Log

B.1. Since draft-ietf-quic-recovery-02

B.2. Since draft-ietf-quic-recovery-01

B.3. Since draft-ietf-quic-recovery-00

B.4. Since draft-iyengar-quic-loss-recovery-01

Authors' Addresses

Jana Iyengar (editor) Google EMail: jri@google.com
Ian Swett (editor) Google EMail: ianswett@google.com
Fork me on GitHub