rfc9002.original   rfc9002.txt 
QUIC J. Iyengar, Ed. Internet Engineering Task Force (IETF) J. Iyengar, Ed.
Internet-Draft Fastly Request for Comments: 9002 Fastly
Intended status: Standards Track I. Swett, Ed. Category: Standards Track I. Swett, Ed.
Expires: 19 July 2021 Google ISSN: 2070-1721 Google
15 January 2021 May 2021
QUIC Loss Detection and Congestion Control QUIC Loss Detection and Congestion Control
draft-ietf-quic-recovery-34
Abstract Abstract
This document describes loss detection and congestion control This document describes loss detection and congestion control
mechanisms for QUIC. mechanisms for QUIC.
Note to Readers
Discussion of this draft takes place on the QUIC working group
mailing list (quic@ietf.org (mailto: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 Status of This Memo
This Internet-Draft is submitted in full conformance with the This is an Internet Standards Track document.
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 https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months This document is a product of the Internet Engineering Task Force
and may be updated, replaced, or obsoleted by other documents at any (IETF). It represents the consensus of the IETF community. It has
time. It is inappropriate to use Internet-Drafts as reference received public review and has been approved for publication by the
material or to cite them other than as "work in progress." Internet Engineering Steering Group (IESG). Further information on
Internet Standards is available in Section 2 of RFC 7841.
This Internet-Draft will expire on 19 July 2021. Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
https://www.rfc-editor.org/info/rfc9002.
Copyright Notice Copyright Notice
Copyright (c) 2021 IETF Trust and the persons identified as the Copyright (c) 2021 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents (https://trustee.ietf.org/ Provisions Relating to IETF Documents
license-info) in effect on the date of publication of this document. (https://trustee.ietf.org/license-info) in effect on the date of
Please review these documents carefully, as they describe your rights publication of this document. Please review these documents
and restrictions with respect to this document. Code Components carefully, as they describe your rights and restrictions with respect
extracted from this document must include Simplified BSD License text to this document. Code Components extracted from this document must
as described in Section 4.e of the Trust Legal Provisions and are include Simplified BSD License text as described in Section 4.e of
provided without warranty as described in the Simplified BSD License. the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 1. Introduction
2. Conventions and Definitions . . . . . . . . . . . . . . . . . 4 2. Conventions and Definitions
3. Design of the QUIC Transmission Machinery . . . . . . . . . . 5 3. Design of the QUIC Transmission Machinery
4. Relevant Differences Between QUIC and TCP . . . . . . . . . . 6 4. Relevant Differences between QUIC and TCP
4.1. Separate Packet Number Spaces . . . . . . . . . . . . . . 6 4.1. Separate Packet Number Spaces
4.2. Monotonically Increasing Packet Numbers . . . . . . . . . 6 4.2. Monotonically Increasing Packet Numbers
4.3. Clearer Loss Epoch . . . . . . . . . . . . . . . . . . . 7 4.3. Clearer Loss Epoch
4.4. No Reneging . . . . . . . . . . . . . . . . . . . . . . . 7 4.4. No Reneging
4.5. More ACK Ranges . . . . . . . . . . . . . . . . . . . . . 7 4.5. More ACK Ranges
4.6. Explicit Correction For Delayed Acknowledgments . . . . . 7 4.6. Explicit Correction for Delayed Acknowledgments
4.7. Probe Timeout Replaces RTO and TLP . . . . . . . . . . . 7 4.7. Probe Timeout Replaces RTO and TLP
4.8. The Minimum Congestion Window is Two Packets . . . . . . 8 4.8. The Minimum Congestion Window Is Two Packets
5. Estimating the Round-Trip Time . . . . . . . . . . . . . . . 8 4.9. Handshake Packets Are Not Special
5.1. Generating RTT samples . . . . . . . . . . . . . . . . . 8 5. Estimating the Round-Trip Time
5.2. Estimating min_rtt . . . . . . . . . . . . . . . . . . . 9 5.1. Generating RTT Samples
5.3. Estimating smoothed_rtt and rttvar . . . . . . . . . . . 10 5.2. Estimating min_rtt
6. Loss Detection . . . . . . . . . . . . . . . . . . . . . . . 12 5.3. Estimating smoothed_rtt and rttvar
6.1. Acknowledgment-Based Detection . . . . . . . . . . . . . 12 6. Loss Detection
6.1.1. Packet Threshold . . . . . . . . . . . . . . . . . . 13 6.1. Acknowledgment-Based Detection
6.1.2. Time Threshold . . . . . . . . . . . . . . . . . . . 13 6.1.1. Packet Threshold
6.2. Probe Timeout . . . . . . . . . . . . . . . . . . . . . . 14 6.1.2. Time Threshold
6.2.1. Computing PTO . . . . . . . . . . . . . . . . . . . . 14 6.2. Probe Timeout
6.2.2. Handshakes and New Paths . . . . . . . . . . . . . . 16 6.2.1. Computing PTO
6.2.3. Speeding Up Handshake Completion . . . . . . . . . . 17 6.2.2. Handshakes and New Paths
6.2.4. Sending Probe Packets . . . . . . . . . . . . . . . . 18 6.2.3. Speeding up Handshake Completion
6.3. Handling Retry Packets . . . . . . . . . . . . . . . . . 19 6.2.4. Sending Probe Packets
6.4. Discarding Keys and Packet State . . . . . . . . . . . . 19 6.3. Handling Retry Packets
7. Congestion Control . . . . . . . . . . . . . . . . . . . . . 20 6.4. Discarding Keys and Packet State
7.1. Explicit Congestion Notification . . . . . . . . . . . . 20 7. Congestion Control
7.2. Initial and Minimum Congestion Window . . . . . . . . . . 21 7.1. Explicit Congestion Notification
7.3. Congestion Control States . . . . . . . . . . . . . . . . 21 7.2. Initial and Minimum Congestion Window
7.3.1. Slow Start . . . . . . . . . . . . . . . . . . . . . 22 7.3. Congestion Control States
7.3.2. Recovery . . . . . . . . . . . . . . . . . . . . . . 22 7.3.1. Slow Start
7.3.3. Congestion Avoidance . . . . . . . . . . . . . . . . 23 7.3.2. Recovery
7.4. Ignoring Loss of Undecryptable Packets . . . . . . . . . 23 7.3.3. Congestion Avoidance
7.5. Probe Timeout . . . . . . . . . . . . . . . . . . . . . . 24 7.4. Ignoring Loss of Undecryptable Packets
7.6. Persistent Congestion . . . . . . . . . . . . . . . . . . 24 7.5. Probe Timeout
7.6.1. Duration . . . . . . . . . . . . . . . . . . . . . . 24 7.6. Persistent Congestion
7.6.2. Establishing Persistent Congestion . . . . . . . . . 25 7.6.1. Duration
7.6.3. Example . . . . . . . . . . . . . . . . . . . . . . . 26 7.6.2. Establishing Persistent Congestion
7.7. Pacing . . . . . . . . . . . . . . . . . . . . . . . . . 27 7.6.3. Example
7.8. Under-utilizing the Congestion Window . . . . . . . . . . 28 7.7. Pacing
8. Security Considerations . . . . . . . . . . . . . . . . . . . 28 7.8. Underutilizing the Congestion Window
8.1. Loss and Congestion Signals . . . . . . . . . . . . . . . 28 8. Security Considerations
8.2. Traffic Analysis . . . . . . . . . . . . . . . . . . . . 28 8.1. Loss and Congestion Signals
8.3. Misreporting ECN Markings . . . . . . . . . . . . . . . . 28 8.2. Traffic Analysis
9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 29 8.3. Misreporting ECN Markings
10. References . . . . . . . . . . . . . . . . . . . . . . . . . 29 9. References
10.1. Normative References . . . . . . . . . . . . . . . . . . 29 9.1. Normative References
10.2. Informative References . . . . . . . . . . . . . . . . . 30 9.2. Informative References
Appendix A. Loss Recovery Pseudocode . . . . . . . . . . . . . . 32 Appendix A. Loss Recovery Pseudocode
A.1. Tracking Sent Packets . . . . . . . . . . . . . . . . . . 32 A.1. Tracking Sent Packets
A.1.1. Sent Packet Fields . . . . . . . . . . . . . . . . . 32 A.1.1. Sent Packet Fields
A.2. Constants of Interest . . . . . . . . . . . . . . . . . . 33 A.2. Constants of Interest
A.3. Variables of interest . . . . . . . . . . . . . . . . . . 33 A.3. Variables of Interest
A.4. Initialization . . . . . . . . . . . . . . . . . . . . . 34 A.4. Initialization
A.5. On Sending a Packet . . . . . . . . . . . . . . . . . . . 34 A.5. On Sending a Packet
A.6. On Receiving a Datagram . . . . . . . . . . . . . . . . . 35 A.6. On Receiving a Datagram
A.7. On Receiving an Acknowledgment . . . . . . . . . . . . . 35 A.7. On Receiving an Acknowledgment
A.8. Setting the Loss Detection Timer . . . . . . . . . . . . 37 A.8. Setting the Loss Detection Timer
A.9. On Timeout . . . . . . . . . . . . . . . . . . . . . . . 39 A.9. On Timeout
A.10. Detecting Lost Packets . . . . . . . . . . . . . . . . . 39 A.10. Detecting Lost Packets
A.11. Upon Dropping Initial or Handshake Keys . . . . . . . . . 40 A.11. Upon Dropping Initial or Handshake Keys
Appendix B. Congestion Control Pseudocode . . . . . . . . . . . 41 Appendix B. Congestion Control Pseudocode
B.1. Constants of interest . . . . . . . . . . . . . . . . . . 41 B.1. Constants of Interest
B.2. Variables of interest . . . . . . . . . . . . . . . . . . 41 B.2. Variables of Interest
B.3. Initialization . . . . . . . . . . . . . . . . . . . . . 42 B.3. Initialization
B.4. On Packet Sent . . . . . . . . . . . . . . . . . . . . . 42 B.4. On Packet Sent
B.5. On Packet Acknowledgment . . . . . . . . . . . . . . . . 42 B.5. On Packet Acknowledgment
B.6. On New Congestion Event . . . . . . . . . . . . . . . . . 43 B.6. On New Congestion Event
B.7. Process ECN Information . . . . . . . . . . . . . . . . . 44 B.7. Process ECN Information
B.8. On Packets Lost . . . . . . . . . . . . . . . . . . . . . 44 B.8. On Packets Lost
B.9. Removing Discarded Packets From Bytes In Flight . . . . . 44 B.9. Removing Discarded Packets from Bytes in Flight
Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 45 Contributors
C.1. Since draft-ietf-quic-recovery-32 . . . . . . . . . . . . 45 Authors' Addresses
C.2. Since draft-ietf-quic-recovery-31 . . . . . . . . . . . . 45
C.3. Since draft-ietf-quic-recovery-30 . . . . . . . . . . . . 45
C.4. Since draft-ietf-quic-recovery-29 . . . . . . . . . . . . 45
C.5. Since draft-ietf-quic-recovery-28 . . . . . . . . . . . . 46
C.6. Since draft-ietf-quic-recovery-27 . . . . . . . . . . . . 46
C.7. Since draft-ietf-quic-recovery-26 . . . . . . . . . . . . 46
C.8. Since draft-ietf-quic-recovery-25 . . . . . . . . . . . . 46
C.9. Since draft-ietf-quic-recovery-24 . . . . . . . . . . . . 46
C.10. Since draft-ietf-quic-recovery-23 . . . . . . . . . . . . 46
C.11. Since draft-ietf-quic-recovery-22 . . . . . . . . . . . . 47
C.12. Since draft-ietf-quic-recovery-21 . . . . . . . . . . . . 47
C.13. Since draft-ietf-quic-recovery-20 . . . . . . . . . . . . 47
C.14. Since draft-ietf-quic-recovery-19 . . . . . . . . . . . . 47
C.15. Since draft-ietf-quic-recovery-18 . . . . . . . . . . . . 48
C.16. Since draft-ietf-quic-recovery-17 . . . . . . . . . . . . 48
C.17. Since draft-ietf-quic-recovery-16 . . . . . . . . . . . . 49
C.18. Since draft-ietf-quic-recovery-14 . . . . . . . . . . . . 49
C.19. Since draft-ietf-quic-recovery-13 . . . . . . . . . . . . 49
C.20. Since draft-ietf-quic-recovery-12 . . . . . . . . . . . . 50
C.21. Since draft-ietf-quic-recovery-11 . . . . . . . . . . . . 50
C.22. Since draft-ietf-quic-recovery-10 . . . . . . . . . . . . 50
C.23. Since draft-ietf-quic-recovery-09 . . . . . . . . . . . . 50
C.24. Since draft-ietf-quic-recovery-08 . . . . . . . . . . . . 50
C.25. Since draft-ietf-quic-recovery-07 . . . . . . . . . . . . 50
C.26. Since draft-ietf-quic-recovery-06 . . . . . . . . . . . . 51
C.27. Since draft-ietf-quic-recovery-05 . . . . . . . . . . . . 51
C.28. Since draft-ietf-quic-recovery-04 . . . . . . . . . . . . 51
C.29. Since draft-ietf-quic-recovery-03 . . . . . . . . . . . . 51
C.30. Since draft-ietf-quic-recovery-02 . . . . . . . . . . . . 51
C.31. Since draft-ietf-quic-recovery-01 . . . . . . . . . . . . 51
C.32. Since draft-ietf-quic-recovery-00 . . . . . . . . . . . . 51
C.33. Since draft-iyengar-quic-loss-recovery-01 . . . . . . . . 51
Appendix D. Contributors . . . . . . . . . . . . . . . . . . . . 52
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 52
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 52
1. Introduction 1. Introduction
QUIC is a secure general-purpose transport protocol, described in QUIC is a secure, general-purpose transport protocol, described in
[QUIC-TRANSPORT]). This document describes loss detection and [QUIC-TRANSPORT]. This document describes loss detection and
congestion control mechanisms for QUIC. congestion control mechanisms for QUIC.
2. Conventions and Definitions 2. Conventions and Definitions
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in "OPTIONAL" in this document are to be interpreted as described in BCP
BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all 14 [RFC2119] [RFC8174] when, and only when, they appear in all
capitals, as shown here. capitals, as shown here.
Definitions of terms that are used in this document: Definitions of terms that are used in this document:
Ack-eliciting frames: All frames other than ACK, PADDING, and Ack-eliciting frames: All frames other than ACK, PADDING, and
CONNECTION_CLOSE are considered ack-eliciting. CONNECTION_CLOSE are considered ack-eliciting.
Ack-eliciting packets: Packets that contain ack-eliciting frames Ack-eliciting packets: Packets that contain ack-eliciting frames
elicit an ACK from the receiver within the maximum acknowledgment elicit an ACK from the receiver within the maximum acknowledgment
delay and are called ack-eliciting packets. delay and are called ack-eliciting packets.
In-flight packets: Packets are considered in-flight when they are In-flight packets: Packets are considered in flight when they are
ack-eliciting or contain a PADDING frame, and they have been sent ack-eliciting or contain a PADDING frame, and they have been sent
but are not acknowledged, declared lost, or discarded along with but are not acknowledged, declared lost, or discarded along with
old keys. old keys.
3. Design of the QUIC Transmission Machinery 3. Design of the QUIC Transmission Machinery
All transmissions in QUIC are sent with a packet-level header, which All transmissions in QUIC are sent with a packet-level header, which
indicates the encryption level and includes a packet sequence number indicates the encryption level and includes a packet sequence number
(referred to below as a packet number). The encryption level (referred to below as a packet number). The encryption level
indicates the packet number space, as described in Section 12.3 in indicates the packet number space, as described in Section 12.3 of
[QUIC-TRANSPORT]. Packet numbers never repeat within a packet number [QUIC-TRANSPORT]. Packet numbers never repeat within a packet number
space for the lifetime of a connection. Packet numbers are sent in space for the lifetime of a connection. Packet numbers are sent in
monotonically increasing order within a space, preventing ambiguity. monotonically increasing order within a space, preventing ambiguity.
It is permitted for some packet numbers to never be used, leaving It is permitted for some packet numbers to never be used, leaving
intentional gaps. intentional gaps.
This design obviates the need for disambiguating between This design obviates the need for disambiguating between
transmissions and retransmissions; this eliminates significant transmissions and retransmissions; this eliminates significant
complexity from QUIC's interpretation of TCP loss detection complexity from QUIC's interpretation of TCP loss detection
mechanisms. mechanisms.
skipping to change at page 5, line 42 skipping to change at line 182
* All packets are acknowledged, though packets that contain no ack- * All packets are acknowledged, though packets that contain no ack-
eliciting frames are only acknowledged along with ack-eliciting eliciting frames are only acknowledged along with ack-eliciting
packets. packets.
* Long header packets that contain CRYPTO frames are critical to the * Long header packets that contain CRYPTO frames are critical to the
performance of the QUIC handshake and use shorter timers for performance of the QUIC handshake and use shorter timers for
acknowledgment. acknowledgment.
* Packets containing frames besides ACK or CONNECTION_CLOSE frames * Packets containing frames besides ACK or CONNECTION_CLOSE frames
count toward congestion control limits and are considered in- count toward congestion control limits and are considered to be in
flight. flight.
* PADDING frames cause packets to contribute toward bytes in flight * PADDING frames cause packets to contribute toward bytes in flight
without directly causing an acknowledgment to be sent. without directly causing an acknowledgment to be sent.
4. Relevant Differences Between QUIC and TCP 4. Relevant Differences between QUIC and TCP
Readers familiar with TCP's loss detection and congestion control Readers familiar with TCP's loss detection and congestion control
will find algorithms here that parallel well-known TCP ones. will find algorithms here that parallel well-known TCP ones.
However, protocol differences between QUIC and TCP contribute to However, protocol differences between QUIC and TCP contribute to
algorithmic differences. These protocol differences are briefly algorithmic differences. These protocol differences are briefly
described below. described below.
4.1. Separate Packet Number Spaces 4.1. Separate Packet Number Spaces
QUIC uses separate packet number spaces for each encryption level, QUIC uses separate packet number spaces for each encryption level,
except 0-RTT and all generations of 1-RTT keys use the same packet except 0-RTT and all generations of 1-RTT keys use the same packet
number space. Separate packet number spaces ensures acknowledgment number space. Separate packet number spaces ensures that the
of packets sent with one level of encryption will not cause spurious acknowledgment of packets sent with one level of encryption will not
retransmission of packets sent with a different encryption level. cause spurious retransmission of packets sent with a different
Congestion control and round-trip time (RTT) measurement are unified encryption level. Congestion control and round-trip time (RTT)
across packet number spaces. measurement are unified across packet number spaces.
4.2. Monotonically Increasing Packet Numbers 4.2. Monotonically Increasing Packet Numbers
TCP conflates transmission order at the sender with delivery order at TCP conflates transmission order at the sender with delivery order at
the receiver, resulting in the retransmission ambiguity problem the receiver, resulting in the retransmission ambiguity problem
([RETRANSMISSION]). QUIC separates transmission order from delivery [RETRANSMISSION]. QUIC separates transmission order from delivery
order: packet numbers indicate transmission order, and delivery order order: packet numbers indicate transmission order, and delivery order
is determined by the stream offsets in STREAM frames. is determined by the stream offsets in STREAM frames.
QUIC's packet number is strictly increasing within a packet number QUIC's packet number is strictly increasing within a packet number
space, and directly encodes transmission order. A higher packet space and directly encodes transmission order. A higher packet
number signifies that the packet was sent later, and a lower packet number signifies that the packet was sent later, and a lower packet
number signifies that the packet was sent earlier. When a packet number signifies that the packet was sent earlier. When a packet
containing ack-eliciting frames is detected lost, QUIC includes containing ack-eliciting frames is detected lost, QUIC includes
necessary frames in a new packet with a new packet number, removing necessary frames in a new packet with a new packet number, removing
ambiguity about which packet is acknowledged when an ACK is received. ambiguity about which packet is acknowledged when an ACK is received.
Consequently, more accurate RTT measurements can be made, spurious Consequently, more accurate RTT measurements can be made, spurious
retransmissions are trivially detected, and mechanisms such as Fast retransmissions are trivially detected, and mechanisms such as Fast
Retransmit can be applied universally, based only on packet number. Retransmit can be applied universally, based only on packet number.
This design point significantly simplifies loss detection mechanisms This design point significantly simplifies loss detection mechanisms
for QUIC. Most TCP mechanisms implicitly attempt to infer for QUIC. Most TCP mechanisms implicitly attempt to infer
transmission ordering based on TCP sequence numbers - a non-trivial transmission ordering based on TCP sequence numbers -- a nontrivial
task, especially when TCP timestamps are not available. task, especially when TCP timestamps are not available.
4.3. Clearer Loss Epoch 4.3. Clearer Loss Epoch
QUIC starts a loss epoch when a packet is lost. The loss epoch ends QUIC starts a loss epoch when a packet is lost. The loss epoch ends
when any packet sent after the start of the epoch is acknowledged. when any packet sent after the start of the epoch is acknowledged.
TCP waits for the gap in the sequence number space to be filled, and TCP waits for the gap in the sequence number space to be filled, and
so if a segment is lost multiple times in a row, the loss epoch may so if a segment is lost multiple times in a row, the loss epoch may
not end for several round trips. Because both should reduce their not end for several round trips. Because both should reduce their
congestion windows only once per epoch, QUIC will do it once for congestion windows only once per epoch, QUIC will do it once for
every round trip that experiences loss, while TCP may only do it once every round trip that experiences loss, while TCP may only do it once
across multiple round trips. across multiple round trips.
4.4. No Reneging 4.4. No Reneging
QUIC ACK frames contain information similar to that in TCP Selective QUIC ACK frames contain information similar to that in TCP Selective
Acknowledgements (SACKs, [RFC2018]). However, QUIC does not allow a Acknowledgments (SACKs) [RFC2018]. However, QUIC does not allow a
packet acknowledgement to be reneged, greatly simplifying packet acknowledgment to be reneged, greatly simplifying
implementations on both sides and reducing memory pressure on the implementations on both sides and reducing memory pressure on the
sender. sender.
4.5. More ACK Ranges 4.5. More ACK Ranges
QUIC supports many ACK ranges, opposed to TCP's 3 SACK ranges. In QUIC supports many ACK ranges, as opposed to TCP's three SACK ranges.
high loss environments, this speeds recovery, reduces spurious In high-loss environments, this speeds recovery, reduces spurious
retransmits, and ensures forward progress without relying on retransmits, and ensures forward progress without relying on
timeouts. timeouts.
4.6. Explicit Correction For Delayed Acknowledgments 4.6. Explicit Correction for Delayed Acknowledgments
QUIC endpoints measure the delay incurred between when a packet is QUIC endpoints measure the delay incurred between when a packet is
received and when the corresponding acknowledgment is sent, allowing received and when the corresponding acknowledgment is sent, allowing
a peer to maintain a more accurate round-trip time estimate; see a peer to maintain a more accurate RTT estimate; see Section 13.2 of
Section 13.2 of [QUIC-TRANSPORT]. [QUIC-TRANSPORT].
4.7. Probe Timeout Replaces RTO and TLP 4.7. Probe Timeout Replaces RTO and TLP
QUIC uses a probe timeout (PTO; see Section 6.2), with a timer based QUIC uses a probe timeout (PTO; see Section 6.2), with a timer based
on TCP's RTO computation; see [RFC6297]. QUIC's PTO includes the on TCP's retransmission timeout (RTO) computation; see [RFC6298].
peer's maximum expected acknowledgment delay instead of using a fixed QUIC's PTO includes the peer's maximum expected acknowledgment delay
minimum timeout. instead of using a fixed minimum timeout.
Similar to the RACK-TLP loss detection algorithm for TCP ([RACK]), Similar to the RACK-TLP loss detection algorithm for TCP [RFC8985],
QUIC does not collapse the congestion window when the PTO expires, QUIC does not collapse the congestion window when the PTO expires,
since a single packet loss at the tail does not indicate persistent since a single packet loss at the tail does not indicate persistent
congestion. Instead, QUIC collapses the congestion window when congestion. Instead, QUIC collapses the congestion window when
persistent congestion is declared; see Section 7.6. In doing this, persistent congestion is declared; see Section 7.6. In doing this,
QUIC avoids unnecessary congestion window reductions, obviating the QUIC avoids unnecessary congestion window reductions, obviating the
need for correcting mechanisms such as F-RTO ([RFC5682]). Since QUIC need for correcting mechanisms such as Forward RTO-Recovery (F-RTO)
does not collapse the congestion window on a PTO expiration, a QUIC [RFC5682]. Since QUIC does not collapse the congestion window on a
sender is not limited from sending more in-flight packets after a PTO PTO expiration, a QUIC sender is not limited from sending more in-
expiration if it still has available congestion window. This occurs flight packets after a PTO expiration if it still has available
when a sender is application-limited and the PTO timer expires. This congestion window. This occurs when a sender is application limited
is more aggressive than TCP's RTO mechanism when application-limited, and the PTO timer expires. This is more aggressive than TCP's RTO
but identical when not application-limited. mechanism when application limited, but identical when not
application limited.
QUIC allows probe packets to temporarily exceed the congestion window QUIC allows probe packets to temporarily exceed the congestion window
whenever the timer expires. whenever the timer expires.
4.8. The Minimum Congestion Window is Two Packets 4.8. The Minimum Congestion Window Is Two Packets
TCP uses a minimum congestion window of one packet. However, loss of TCP uses a minimum congestion window of one packet. However, loss of
that single packet means that the sender needs to waiting for a PTO that single packet means that the sender needs to wait for a PTO to
(Section 6.2) to recover, which can be much longer than a round-trip recover (Section 6.2), which can be much longer than an RTT. Sending
time. Sending a single ack-eliciting packet also increases the a single ack-eliciting packet also increases the chances of incurring
chances of incurring additional latency when a receiver delays its additional latency when a receiver delays its acknowledgment.
acknowledgment.
QUIC therefore recommends that the minimum congestion window be two QUIC therefore recommends that the minimum congestion window be two
packets. While this increases network load, it is considered safe, packets. While this increases network load, it is considered safe
since the sender will still reduce its sending rate exponentially since the sender will still reduce its sending rate exponentially
under persistent congestion (Section 6.2). under persistent congestion (Section 6.2).
4.9. Handshake Packets Are Not Special
TCP treats the loss of SYN or SYN-ACK packet as persistent congestion
and reduces the congestion window to one packet; see [RFC5681]. QUIC
treats loss of a packet containing handshake data the same as other
losses.
5. Estimating the Round-Trip Time 5. Estimating the Round-Trip Time
At a high level, an endpoint measures the time from when a packet was At a high level, an endpoint measures the time from when a packet was
sent to when it is acknowledged as a round-trip time (RTT) sample. sent to when it is acknowledged as an RTT sample. The endpoint uses
The endpoint uses RTT samples and peer-reported host delays (see RTT samples and peer-reported host delays (see Section 13.2 of
Section 13.2 of [QUIC-TRANSPORT]) to generate a statistical [QUIC-TRANSPORT]) to generate a statistical description of the
description of the network path's RTT. An endpoint computes the network path's RTT. An endpoint computes the following three values
following three values for each path: the minimum value over a period for each path: the minimum value over a period of time (min_rtt), an
of time (min_rtt), an exponentially-weighted moving average exponentially weighted moving average (smoothed_rtt), and the mean
(smoothed_rtt), and the mean deviation (referred to as "variation" in deviation (referred to as "variation" in the rest of this document)
the rest of this document) in the observed RTT samples (rttvar). in the observed RTT samples (rttvar).
5.1. Generating RTT samples 5.1. Generating RTT Samples
An endpoint generates an RTT sample on receiving an ACK frame that An endpoint generates an RTT sample on receiving an ACK frame that
meets the following two conditions: meets the following two conditions:
* the largest acknowledged packet number is newly acknowledged, and * the largest acknowledged packet number is newly acknowledged, and
* at least one of the newly acknowledged packets was ack-eliciting. * at least one of the newly acknowledged packets was ack-eliciting.
The RTT sample, latest_rtt, is generated as the time elapsed since The RTT sample, latest_rtt, is generated as the time elapsed since
the largest acknowledged packet was sent: the largest acknowledged packet was sent:
skipping to change at page 9, line 18 skipping to change at line 348
the RTT sample measurement, it is used to adjust the RTT sample in the RTT sample measurement, it is used to adjust the RTT sample in
subsequent computations of smoothed_rtt and rttvar (Section 5.3). subsequent computations of smoothed_rtt and rttvar (Section 5.3).
To avoid generating multiple RTT samples for a single packet, an ACK To avoid generating multiple RTT samples for a single packet, an ACK
frame SHOULD NOT be used to update RTT estimates if it does not newly frame SHOULD NOT be used to update RTT estimates if it does not newly
acknowledge the largest acknowledged packet. acknowledge the largest acknowledged packet.
An RTT sample MUST NOT be generated on receiving an ACK frame that An RTT sample MUST NOT be generated on receiving an ACK frame that
does not newly acknowledge at least one ack-eliciting packet. A peer does not newly acknowledge at least one ack-eliciting packet. A peer
usually does not send an ACK frame when only non-ack-eliciting usually does not send an ACK frame when only non-ack-eliciting
packets are received. Therefore an ACK frame that contains packets are received. Therefore, an ACK frame that contains
acknowledgments for only non-ack-eliciting packets could include an acknowledgments for only non-ack-eliciting packets could include an
arbitrarily large ACK Delay value. Ignoring such ACK frames avoids arbitrarily large ACK Delay value. Ignoring such ACK frames avoids
complications in subsequent smoothed_rtt and rttvar computations. complications in subsequent smoothed_rtt and rttvar computations.
A sender might generate multiple RTT samples per RTT when multiple A sender might generate multiple RTT samples per RTT when multiple
ACK frames are received within an RTT. As suggested in [RFC6298], ACK frames are received within an RTT. As suggested in [RFC6298],
doing so might result in inadequate history in smoothed_rtt and doing so might result in inadequate history in smoothed_rtt and
rttvar. Ensuring that RTT estimates retain sufficient history is an rttvar. Ensuring that RTT estimates retain sufficient history is an
open research question. open research question.
5.2. Estimating min_rtt 5.2. Estimating min_rtt
min_rtt is the sender's estimate of the minimum RTT observed for a min_rtt is the sender's estimate of the minimum RTT observed for a
given network path over a period of time. In this document, min_rtt given network path over a period of time. In this document, min_rtt
is used by loss detection to reject implausibly small rtt samples. is used by loss detection to reject implausibly small RTT samples.
min_rtt MUST be set to the latest_rtt on the first RTT sample. min_rtt MUST be set to the latest_rtt on the first RTT sample.
min_rtt MUST be set to the lesser of min_rtt and latest_rtt min_rtt MUST be set to the lesser of min_rtt and latest_rtt
(Section 5.1) on all other samples. (Section 5.1) on all other samples.
An endpoint uses only locally observed times in computing the min_rtt An endpoint uses only locally observed times in computing the min_rtt
and does not adjust for acknowledgment delays reported by the peer. and does not adjust for acknowledgment delays reported by the peer.
Doing so allows the endpoint to set a lower bound for the Doing so allows the endpoint to set a lower bound for the
smoothed_rtt based entirely on what it observes (see Section 5.3), smoothed_rtt based entirely on what it observes (see Section 5.3) and
and limits potential underestimation due to erroneously-reported limits potential underestimation due to erroneously reported delays
delays by the peer. by the peer.
The RTT for a network path may change over time. If a path's actual The RTT for a network path may change over time. If a path's actual
RTT decreases, the min_rtt will adapt immediately on the first low RTT decreases, the min_rtt will adapt immediately on the first low
sample. If the path's actual RTT increases however, the min_rtt will sample. If the path's actual RTT increases, however, the min_rtt
not adapt to it, allowing future RTT samples that are smaller than will not adapt to it, allowing future RTT samples that are smaller
the new RTT to be included in smoothed_rtt. than the new RTT to be included in smoothed_rtt.
Endpoints SHOULD set the min_rtt to the newest RTT sample after Endpoints SHOULD set the min_rtt to the newest RTT sample after
persistent congestion is established. This is to allow a connection persistent congestion is established. This avoids repeatedly
to reset its estimate of min_rtt and smoothed_rtt (Section 5.3) after declaring persistent congestion when the RTT increases. This also
a disruptive network event, and because it is possible that an allows a connection to reset its estimate of min_rtt and smoothed_rtt
increase in path delay resulted in persistent congestion being after a disruptive network event; see Section 5.3.
incorrectly declared.
Endpoints MAY re-establish the min_rtt at other times in the Endpoints MAY reestablish the min_rtt at other times in the
connection, such as when traffic volume is low and an acknowledgment connection, such as when traffic volume is low and an acknowledgment
is received with a low acknowledgment delay. Implementations SHOULD is received with a low acknowledgment delay. Implementations SHOULD
NOT refresh the min_rtt value too often, since the actual minimum RTT NOT refresh the min_rtt value too often since the actual minimum RTT
of the path is not frequently observable. of the path is not frequently observable.
5.3. Estimating smoothed_rtt and rttvar 5.3. Estimating smoothed_rtt and rttvar
smoothed_rtt is an exponentially-weighted moving average of an smoothed_rtt is an exponentially weighted moving average of an
endpoint's RTT samples, and rttvar estimates the variation in the RTT endpoint's RTT samples, and rttvar estimates the variation in the RTT
samples using a mean variation. samples using a mean variation.
The calculation of smoothed_rtt uses RTT samples after adjusting them The calculation of smoothed_rtt uses RTT samples after adjusting them
for acknowledgment delays. These delays are decoded from the ACK for acknowledgment delays. These delays are decoded from the ACK
Delay field of ACK frames as described in Section 19.3 of Delay field of ACK frames as described in Section 19.3 of
[QUIC-TRANSPORT]. [QUIC-TRANSPORT].
The peer might report acknowledgment delays that are larger than the The peer might report acknowledgment delays that are larger than the
peer's max_ack_delay during the handshake (Section 13.2.1 of peer's max_ack_delay during the handshake (Section 13.2.1 of
[QUIC-TRANSPORT]). To account for this, the endpoint SHOULD ignore [QUIC-TRANSPORT]). To account for this, the endpoint SHOULD ignore
max_ack_delay until the handshake is confirmed, as defined in max_ack_delay until the handshake is confirmed, as defined in
Section 4.1.2 of [QUIC-TLS]. When they occur, these large Section 4.1.2 of [QUIC-TLS]. When they occur, these large
acknowledgment delays are likely to be non-repeating and limited to acknowledgment delays are likely to be non-repeating and limited to
the handshake. The endpoint can therefore use them without limiting the handshake. The endpoint can therefore use them without limiting
them to the max_ack_delay, avoiding unnecessary inflation of the RTT them to the max_ack_delay, avoiding unnecessary inflation of the RTT
estimate. estimate.
Note that a large acknowledgment delay can result in a substantially Note that a large acknowledgment delay can result in a substantially
inflated smoothed_rtt, if there is either an error in the peer's inflated smoothed_rtt if there is an error either in the peer's
reporting of the acknowledgment delay or in the endpoint's min_rtt reporting of the acknowledgment delay or in the endpoint's min_rtt
estimate. Therefore, prior to handshake confirmation, an endpoint estimate. Therefore, prior to handshake confirmation, an endpoint
MAY ignore RTT samples if adjusting the RTT sample for acknowledgment MAY ignore RTT samples if adjusting the RTT sample for acknowledgment
delay causes the sample to be less than the min_rtt. delay causes the sample to be less than the min_rtt.
After the handshake is confirmed, any acknowledgment delays reported After the handshake is confirmed, any acknowledgment delays reported
by the peer that are greater than the peer's max_ack_delay are by the peer that are greater than the peer's max_ack_delay are
attributed to unintentional but potentially repeating delays, such as attributed to unintentional but potentially repeating delays, such as
scheduler latency at the peer or loss of previous acknowledgments. scheduler latency at the peer or loss of previous acknowledgments.
Excess delays could also be due to a non-compliant receiver. Excess delays could also be due to a noncompliant receiver.
Therefore, these extra delays are considered effectively part of path Therefore, these extra delays are considered effectively part of path
delay and incorporated into the RTT estimate. delay and incorporated into the RTT estimate.
Therefore, when adjusting an RTT sample using peer-reported Therefore, when adjusting an RTT sample using peer-reported
acknowledgment delays, an endpoint: acknowledgment delays, an endpoint:
* MAY ignore the acknowledgment delay for Initial packets, since * MAY ignore the acknowledgment delay for Initial packets, since
these acknowledgments are not delayed by the peer (Section 13.2.1 these acknowledgments are not delayed by the peer (Section 13.2.1
of [QUIC-TRANSPORT]); of [QUIC-TRANSPORT]);
skipping to change at page 11, line 49 skipping to change at line 474
smoothed_rtt and rttvar are initialized as follows, where kInitialRtt smoothed_rtt and rttvar are initialized as follows, where kInitialRtt
contains the initial RTT value: contains the initial RTT value:
smoothed_rtt = kInitialRtt smoothed_rtt = kInitialRtt
rttvar = kInitialRtt / 2 rttvar = kInitialRtt / 2
RTT samples for the network path are recorded in latest_rtt; see RTT samples for the network path are recorded in latest_rtt; see
Section 5.1. On the first RTT sample after initialization, the Section 5.1. On the first RTT sample after initialization, the
estimator is reset using that sample. This ensures that the estimator is reset using that sample. This ensures that the
estimator retains no history of past samples. estimator retains no history of past samples. Packets sent on other
paths do not contribute RTT samples to the current path, as described
in Section 9.4 of [QUIC-TRANSPORT].
On the first RTT sample after initialization, smoothed_rtt and rttvar On the first RTT sample after initialization, smoothed_rtt and rttvar
are set as follows: are set as follows:
smoothed_rtt = latest_rtt smoothed_rtt = latest_rtt
rttvar = latest_rtt / 2 rttvar = latest_rtt / 2
On subsequent RTT samples, smoothed_rtt and rttvar evolve as follows: On subsequent RTT samples, smoothed_rtt and rttvar evolve as follows:
ack_delay = decoded acknowledgment delay from ACK frame ack_delay = decoded acknowledgment delay from ACK frame
if (handshake confirmed): if (handshake confirmed):
ack_delay = min(ack_delay, max_ack_delay) ack_delay = min(ack_delay, max_ack_delay)
adjusted_rtt = latest_rtt adjusted_rtt = latest_rtt
if (min_rtt + ack_delay < latest_rtt): if (latest_rtt >= min_rtt + ack_delay):
adjusted_rtt = latest_rtt - ack_delay adjusted_rtt = latest_rtt - ack_delay
smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * adjusted_rtt smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * adjusted_rtt
rttvar_sample = abs(smoothed_rtt - adjusted_rtt) rttvar_sample = abs(smoothed_rtt - adjusted_rtt)
rttvar = 3/4 * rttvar + 1/4 * rttvar_sample rttvar = 3/4 * rttvar + 1/4 * rttvar_sample
6. Loss Detection 6. Loss Detection
QUIC senders use acknowledgments to detect lost packets, and a probe QUIC senders use acknowledgments to detect lost packets and a PTO to
time out (see Section 6.2) to ensure acknowledgments are received. ensure acknowledgments are received; see Section 6.2. This section
This section provides a description of these algorithms. provides a description of these algorithms.
If a packet is lost, the QUIC transport needs to recover from that If a packet is lost, the QUIC transport needs to recover from that
loss, such as by retransmitting the data, sending an updated frame, loss, such as by retransmitting the data, sending an updated frame,
or discarding the frame. For more information, see Section 13.3 of or discarding the frame. For more information, see Section 13.3 of
[QUIC-TRANSPORT]. [QUIC-TRANSPORT].
Loss detection is separate per packet number space, unlike RTT Loss detection is separate per packet number space, unlike RTT
measurement and congestion control, because RTT and congestion measurement and congestion control, because RTT and congestion
control are properties of the path, whereas loss detection also control are properties of the path, whereas loss detection also
relies upon key availability. relies upon key availability.
6.1. Acknowledgment-Based Detection 6.1. Acknowledgment-Based Detection
Acknowledgment-based loss detection implements the spirit of TCP's Acknowledgment-based loss detection implements the spirit of TCP's
Fast Retransmit ([RFC5681]), Early Retransmit ([RFC5827]), FACK Fast Retransmit [RFC5681], Early Retransmit [RFC5827], Forward
([FACK]), SACK loss recovery ([RFC6675]), and RACK-TLP ([RACK]). Acknowledgment [FACK], SACK loss recovery [RFC6675], and RACK-TLP
This section provides an overview of how these algorithms are [RFC8985]. This section provides an overview of how these algorithms
implemented in QUIC. are implemented in QUIC.
A packet is declared lost if it meets all the following conditions: A packet is declared lost if it meets all of the following
conditions:
* The packet is unacknowledged, in-flight, and was sent prior to an * The packet is unacknowledged, in flight, and was sent prior to an
acknowledged packet. acknowledged packet.
* The packet was sent kPacketThreshold packets before an * The packet was sent kPacketThreshold packets before an
acknowledged packet (Section 6.1.1), or it was sent long enough in acknowledged packet (Section 6.1.1), or it was sent long enough in
the past (Section 6.1.2). the past (Section 6.1.2).
The acknowledgment indicates that a packet sent later was delivered, The acknowledgment indicates that a packet sent later was delivered,
and the packet and time thresholds provide some tolerance for packet and the packet and time thresholds provide some tolerance for packet
reordering. reordering.
Spuriously declaring packets as lost leads to unnecessary Spuriously declaring packets as lost leads to unnecessary
retransmissions and may result in degraded performance due to the retransmissions and may result in degraded performance due to the
actions of the congestion controller upon detecting loss. actions of the congestion controller upon detecting loss.
Implementations can detect spurious retransmissions and increase the Implementations can detect spurious retransmissions and increase the
reordering threshold in packets or time to reduce future spurious packet or time reordering threshold to reduce future spurious
retransmissions and loss events. Implementations with adaptive time retransmissions and loss events. Implementations with adaptive time
thresholds MAY choose to start with smaller initial reordering thresholds MAY choose to start with smaller initial reordering
thresholds to minimize recovery latency. thresholds to minimize recovery latency.
6.1.1. Packet Threshold 6.1.1. Packet Threshold
The RECOMMENDED initial value for the packet reordering threshold The RECOMMENDED initial value for the packet reordering threshold
(kPacketThreshold) is 3, based on best practices for TCP loss (kPacketThreshold) is 3, based on best practices for TCP loss
detection ([RFC5681], [RFC6675]). In order to remain similar to TCP, detection [RFC5681] [RFC6675]. In order to remain similar to TCP,
implementations SHOULD NOT use a packet threshold less than 3; see implementations SHOULD NOT use a packet threshold less than 3; see
[RFC5681]. [RFC5681].
Some networks may exhibit higher degrees of packet reordering, Some networks may exhibit higher degrees of packet reordering,
causing a sender to detect spurious losses. Additionally, packet causing a sender to detect spurious losses. Additionally, packet
reordering could be more common with QUIC than TCP, because network reordering could be more common with QUIC than TCP because network
elements that could observe and reorder TCP packets cannot do that elements that could observe and reorder TCP packets cannot do that
for QUIC, because QUIC packet numbers are encrypted. Algorithms that for QUIC and also because QUIC packet numbers are encrypted.
increase the reordering threshold after spuriously detecting losses, Algorithms that increase the reordering threshold after spuriously
such as RACK [RACK], have proven to be useful in TCP and are expected detecting losses, such as RACK [RFC8985], have proven to be useful in
to be at least as useful in QUIC. TCP and are expected to be at least as useful in QUIC.
6.1.2. Time Threshold 6.1.2. Time Threshold
Once a later packet within the same packet number space has been Once a later packet within the same packet number space has been
acknowledged, an endpoint SHOULD declare an earlier packet lost if it acknowledged, an endpoint SHOULD declare an earlier packet lost if it
was sent a threshold amount of time in the past. To avoid declaring was sent a threshold amount of time in the past. To avoid declaring
packets as lost too early, this time threshold MUST be set to at packets as lost too early, this time threshold MUST be set to at
least the local timer granularity, as indicated by the kGranularity least the local timer granularity, as indicated by the kGranularity
constant. The time threshold is: constant. The time threshold is:
skipping to change at page 14, line 12 skipping to change at line 584
Using max(smoothed_rtt, latest_rtt) protects from the two following Using max(smoothed_rtt, latest_rtt) protects from the two following
cases: cases:
* the latest RTT sample is lower than the smoothed RTT, perhaps due * the latest RTT sample is lower than the smoothed RTT, perhaps due
to reordering where the acknowledgment encountered a shorter path; to reordering where the acknowledgment encountered a shorter path;
* the latest RTT sample is higher than the smoothed RTT, perhaps due * the latest RTT sample is higher than the smoothed RTT, perhaps due
to a sustained increase in the actual RTT, but the smoothed RTT to a sustained increase in the actual RTT, but the smoothed RTT
has not yet caught up. has not yet caught up.
The RECOMMENDED time threshold (kTimeThreshold), expressed as a The RECOMMENDED time threshold (kTimeThreshold), expressed as an RTT
round-trip time multiplier, is 9/8. The RECOMMENDED value of the multiplier, is 9/8. The RECOMMENDED value of the timer granularity
timer granularity (kGranularity) is 1ms. (kGranularity) is 1 millisecond.
Note: TCP's RACK ([RACK]) specifies a slightly larger threshold, | Note: TCP's RACK [RFC8985] specifies a slightly larger
equivalent to 5/4, for a similar purpose. Experience with QUIC | threshold, equivalent to 5/4, for a similar purpose.
shows that 9/8 works well. | Experience with QUIC shows that 9/8 works well.
Implementations MAY experiment with absolute thresholds, thresholds Implementations MAY experiment with absolute thresholds, thresholds
from previous connections, adaptive thresholds, or including RTT from previous connections, adaptive thresholds, or the including of
variation. Smaller thresholds reduce reordering resilience and RTT variation. Smaller thresholds reduce reordering resilience and
increase spurious retransmissions, and larger thresholds increase increase spurious retransmissions, and larger thresholds increase
loss detection delay. loss detection delay.
6.2. Probe Timeout 6.2. Probe Timeout
A Probe Timeout (PTO) triggers sending one or two probe datagrams A Probe Timeout (PTO) triggers the sending of one or two probe
when ack-eliciting packets are not acknowledged within the expected datagrams when ack-eliciting packets are not acknowledged within the
period of time or the server may not have validated the client's expected period of time or the server may not have validated the
address. A PTO enables a connection to recover from loss of tail client's address. A PTO enables a connection to recover from loss of
packets or acknowledgments. tail packets or acknowledgments.
As with loss detection, the probe timeout is per packet number space. As with loss detection, the PTO is per packet number space. That is,
That is, a PTO value is computed per packet number space. a PTO value is computed per packet number space.
A PTO timer expiration event does not indicate packet loss and MUST A PTO timer expiration event does not indicate packet loss and MUST
NOT cause prior unacknowledged packets to be marked as lost. When an NOT cause prior unacknowledged packets to be marked as lost. When an
acknowledgment is received that newly acknowledges packets, loss acknowledgment is received that newly acknowledges packets, loss
detection proceeds as dictated by packet and time threshold detection proceeds as dictated by the packet and time threshold
mechanisms; see Section 6.1. mechanisms; see Section 6.1.
The PTO algorithm used in QUIC implements the reliability functions The PTO algorithm used in QUIC implements the reliability functions
of Tail Loss Probe [RACK], RTO [RFC5681], and F-RTO algorithms for of Tail Loss Probe [RFC8985], RTO [RFC5681], and F-RTO algorithms for
TCP [RFC5682]. The timeout computation is based on TCP's TCP [RFC5682]. The timeout computation is based on TCP's RTO period
retransmission timeout period [RFC6298]. [RFC6298].
6.2.1. Computing PTO 6.2.1. Computing PTO
When an ack-eliciting packet is transmitted, the sender schedules a When an ack-eliciting packet is transmitted, the sender schedules a
timer for the PTO period as follows: timer for the PTO period as follows:
PTO = smoothed_rtt + max(4*rttvar, kGranularity) + max_ack_delay PTO = smoothed_rtt + max(4*rttvar, kGranularity) + max_ack_delay
The PTO period is the amount of time that a sender ought to wait for The PTO period is the amount of time that a sender ought to wait for
an acknowledgment of a sent packet. This time period includes the an acknowledgment of a sent packet. This time period includes the
estimated network roundtrip-time (smoothed_rtt), the variation in the estimated network RTT (smoothed_rtt), the variation in the estimate
estimate (4*rttvar), and max_ack_delay, to account for the maximum (4*rttvar), and max_ack_delay, to account for the maximum time by
time by which a receiver might delay sending an acknowledgment. which a receiver might delay sending an acknowledgment.
When the PTO is armed for Initial or Handshake packet number spaces, When the PTO is armed for Initial or Handshake packet number spaces,
the max_ack_delay in the PTO period computation is set to 0, since the max_ack_delay in the PTO period computation is set to 0, since
the peer is expected to not delay these packets intentionally; see the peer is expected to not delay these packets intentionally; see
13.2.1 of [QUIC-TRANSPORT]. Section 13.2.1 of [QUIC-TRANSPORT].
The PTO period MUST be at least kGranularity, to avoid the timer The PTO period MUST be at least kGranularity to avoid the timer
expiring immediately. expiring immediately.
When ack-eliciting packets in multiple packet number spaces are in When ack-eliciting packets in multiple packet number spaces are in
flight, the timer MUST be set to the earlier value of the Initial and flight, the timer MUST be set to the earlier value of the Initial and
Handshake packet number spaces. Handshake packet number spaces.
An endpoint MUST NOT set its PTO timer for the application data An endpoint MUST NOT set its PTO timer for the Application Data
packet number space until the handshake is confirmed. Doing so packet number space until the handshake is confirmed. Doing so
prevents the endpoint from retransmitting information in packets when prevents the endpoint from retransmitting information in packets when
either the peer does not yet have the keys to process them or the either the peer does not yet have the keys to process them or the
endpoint does not yet have the keys to process their acknowledgments. endpoint does not yet have the keys to process their acknowledgments.
For example, this can happen when a client sends 0-RTT packets to the For example, this can happen when a client sends 0-RTT packets to the
server; it does so without knowing whether the server will be able to server; it does so without knowing whether the server will be able to
decrypt them. Similarly, this can happen when a server sends 1-RTT decrypt them. Similarly, this can happen when a server sends 1-RTT
packets before confirming that the client has verified the server's packets before confirming that the client has verified the server's
certificate and can therefore read these 1-RTT packets. certificate and can therefore read these 1-RTT packets.
A sender SHOULD restart its PTO timer every time an ack-eliciting A sender SHOULD restart its PTO timer every time an ack-eliciting
packet is sent or acknowledged, or when Initial or Handshake keys are packet is sent or acknowledged, or when Initial or Handshake keys are
discarded (Section 4.9 of [QUIC-TLS]). This ensures the PTO is discarded (Section 4.9 of [QUIC-TLS]). This ensures the PTO is
always set based on the latest estimate of the round-trip time and always set based on the latest estimate of the RTT and for the
for the correct packet across packet number spaces. correct packet across packet number spaces.
When a PTO timer expires, the PTO backoff MUST be increased, When a PTO timer expires, the PTO backoff MUST be increased,
resulting in the PTO period being set to twice its current value. resulting in the PTO period being set to twice its current value.
The PTO backoff factor is reset when an acknowledgment is received, The PTO backoff factor is reset when an acknowledgment is received,
except in the following case. A server might take longer to respond except in the following case. A server might take longer to respond
to packets during the handshake than otherwise. To protect such a to packets during the handshake than otherwise. To protect such a
server from repeated client probes, the PTO backoff is not reset at a server from repeated client probes, the PTO backoff is not reset at a
client that is not yet certain that the server has finished client that is not yet certain that the server has finished
validating the client's address. That is, a client does not reset validating the client's address. That is, a client does not reset
the PTO backoff factor on receiving acknowledgments in Initial the PTO backoff factor on receiving acknowledgments in Initial
packets. packets.
This exponential reduction in the sender's rate is important because This exponential reduction in the sender's rate is important because
consecutive PTOs might be caused by loss of packets or consecutive PTOs might be caused by loss of packets or
acknowledgments due to severe congestion. Even when there are ack- acknowledgments due to severe congestion. Even when there are ack-
eliciting packets in-flight in multiple packet number spaces, the eliciting packets in flight in multiple packet number spaces, the
exponential increase in probe timeout occurs across all spaces to exponential increase in PTO occurs across all spaces to prevent
prevent excess load on the network. For example, a timeout in the excess load on the network. For example, a timeout in the Initial
Initial packet number space doubles the length of the timeout in the packet number space doubles the length of the timeout in the
Handshake packet number space. Handshake packet number space.
The total length of time over which consecutive PTOs expire is The total length of time over which consecutive PTOs expire is
limited by the idle timeout. limited by the idle timeout.
The PTO timer MUST NOT be set if a timer is set for time threshold The PTO timer MUST NOT be set if a timer is set for time threshold
loss detection; see Section 6.1.2. A timer that is set for time loss detection; see Section 6.1.2. A timer that is set for time
threshold loss detection will expire earlier than the PTO timer in threshold loss detection will expire earlier than the PTO timer in
most cases and is less likely to spuriously retransmit data. most cases and is less likely to spuriously retransmit data.
6.2.2. Handshakes and New Paths 6.2.2. Handshakes and New Paths
Resumed connections over the same network MAY use the previous Resumed connections over the same network MAY use the previous
connection's final smoothed RTT value as the resumed connection's connection's final smoothed RTT value as the resumed connection's
initial RTT. When no previous RTT is available, the initial RTT initial RTT. When no previous RTT is available, the initial RTT
SHOULD be set to 333ms. This results in handshakes starting with a SHOULD be set to 333 milliseconds. This results in handshakes
PTO of 1 second, as recommended for TCP's initial retransmission starting with a PTO of 1 second, as recommended for TCP's initial
timeout; see Section 2 of [RFC6298]. RTO; see Section 2 of [RFC6298].
A connection MAY use the delay between sending a PATH_CHALLENGE and A connection MAY use the delay between sending a PATH_CHALLENGE and
receiving a PATH_RESPONSE to set the initial RTT (see kInitialRtt in receiving a PATH_RESPONSE to set the initial RTT (see kInitialRtt in
Appendix A.2) for a new path, but the delay SHOULD NOT be considered Appendix A.2) for a new path, but the delay SHOULD NOT be considered
an RTT sample. an RTT sample.
Initial packets and Handshake packets could be never acknowledged, When the Initial keys and Handshake keys are discarded (see
but they are removed from bytes in flight when the Initial and Section 6.4), any Initial packets and Handshake packets can no longer
Handshake keys are discarded, as described below in Section 6.4. be acknowledged, so they are removed from bytes in flight. When
When Initial or Handshake keys are discarded, the PTO and loss Initial or Handshake keys are discarded, the PTO and loss detection
detection timers MUST be reset, because discarding keys indicates timers MUST be reset, because discarding keys indicates forward
forward progress and the loss detection timer might have been set for progress and the loss detection timer might have been set for a now-
a now discarded packet number space. discarded packet number space.
6.2.2.1. Before Address Validation 6.2.2.1. Before Address Validation
Until the server has validated the client's address on the path, the Until the server has validated the client's address on the path, the
amount of data it can send is limited to three times the amount of amount of data it can send is limited to three times the amount of
data received, as specified in Section 8.1 of [QUIC-TRANSPORT]. If data received, as specified in Section 8.1 of [QUIC-TRANSPORT]. If
no additional data can be sent, the server's PTO timer MUST NOT be no additional data can be sent, the server's PTO timer MUST NOT be
armed until datagrams have been received from the client, because armed until datagrams have been received from the client because
packets sent on PTO count against the anti-amplification limit. Note packets sent on PTO count against the anti-amplification limit.
that the server could fail to validate the client's address even if
0-RTT is accepted. When the server receives a datagram from the client, the
amplification limit is increased and the server resets the PTO timer.
If the PTO timer is then set to a time in the past, it is executed
immediately. Doing so avoids sending new 1-RTT packets prior to
packets critical to the completion of the handshake. In particular,
this can happen when 0-RTT is accepted but the server fails to
validate the client's address.
Since the server could be blocked until more datagrams are received Since the server could be blocked until more datagrams are received
from the client, it is the client's responsibility to send packets to from the client, it is the client's responsibility to send packets to
unblock the server until it is certain that the server has finished unblock the server until it is certain that the server has finished
its address validation (see Section 8 of [QUIC-TRANSPORT]). That is, its address validation (see Section 8 of [QUIC-TRANSPORT]). That is,
the client MUST set the probe timer if the client has not received an the client MUST set the PTO timer if the client has not received an
acknowledgment for any of its Handshake packets and the handshake is acknowledgment for any of its Handshake packets and the handshake is
not confirmed (see Section 4.1.2 of [QUIC-TLS]), even if there are no not confirmed (see Section 4.1.2 of [QUIC-TLS]), even if there are no
packets in flight. When the PTO fires, the client MUST send a packets in flight. When the PTO fires, the client MUST send a
Handshake packet if it has Handshake keys, otherwise it MUST send an Handshake packet if it has Handshake keys, otherwise it MUST send an
Initial packet in a UDP datagram with a payload of at least 1200 Initial packet in a UDP datagram with a payload of at least 1200
bytes. bytes.
6.2.3. Speeding Up Handshake Completion 6.2.3. Speeding up Handshake Completion
When a server receives an Initial packet containing duplicate CRYPTO When a server receives an Initial packet containing duplicate CRYPTO
data, it can assume the client did not receive all of the server's data, it can assume the client did not receive all of the server's
CRYPTO data sent in Initial packets, or the client's estimated RTT is CRYPTO data sent in Initial packets, or the client's estimated RTT is
too small. When a client receives Handshake or 1-RTT packets prior too small. When a client receives Handshake or 1-RTT packets prior
to obtaining Handshake keys, it may assume some or all of the to obtaining Handshake keys, it may assume some or all of the
server's Initial packets were lost. server's Initial packets were lost.
To speed up handshake completion under these conditions, an endpoint To speed up handshake completion under these conditions, an endpoint
MAY, for a limited number of times per connection, send a packet MAY, for a limited number of times per connection, send a packet
containing unacknowledged CRYPTO data earlier than the PTO expiry, containing unacknowledged CRYPTO data earlier than the PTO expiry,
subject to the address validation limits in Section 8.1 of subject to the address validation limits in Section 8.1 of
[QUIC-TRANSPORT]. Doing so at most once for each connection is [QUIC-TRANSPORT]. Doing so at most once for each connection is
adequate to quickly recover from a single packet loss. An endpoint adequate to quickly recover from a single packet loss. An endpoint
that always retransmits packets in response to receiving packets that that always retransmits packets in response to receiving packets that
it cannot process risks creating an infinite exchange of packets. it cannot process risks creating an infinite exchange of packets.
Endpoints can also use coalesced packets (see Section 12.2 of Endpoints can also use coalesced packets (see Section 12.2 of
[QUIC-TRANSPORT]) to ensure that each datagram elicits at least one [QUIC-TRANSPORT]) to ensure that each datagram elicits at least one
acknowledgment. For example, a client can coalesce an Initial packet acknowledgment. For example, a client can coalesce an Initial packet
containing PING and PADDING frames with a 0-RTT data packet and a containing PING and PADDING frames with a 0-RTT data packet, and a
server can coalesce an Initial packet containing a PING frame with server can coalesce an Initial packet containing a PING frame with
one or more packets in its first flight. one or more packets in its first flight.
6.2.4. Sending Probe Packets 6.2.4. Sending Probe Packets
When a PTO timer expires, a sender MUST send at least one ack- When a PTO timer expires, a sender MUST send at least one ack-
eliciting packet in the packet number space as a probe. An endpoint eliciting packet in the packet number space as a probe. An endpoint
MAY send up to two full-sized datagrams containing ack-eliciting MAY send up to two full-sized datagrams containing ack-eliciting
packets, to avoid an expensive consecutive PTO expiration due to a packets to avoid an expensive consecutive PTO expiration due to a
single lost datagram, or transmit data from multiple packet number single lost datagram or to transmit data from multiple packet number
spaces. All probe packets sent on a PTO MUST be ack-eliciting. spaces. All probe packets sent on a PTO MUST be ack-eliciting.
In addition to sending data in the packet number space for which the In addition to sending data in the packet number space for which the
timer expired, the sender SHOULD send ack-eliciting packets from timer expired, the sender SHOULD send ack-eliciting packets from
other packet number spaces with in-flight data, coalescing packets if other packet number spaces with in-flight data, coalescing packets if
possible. This is particularly valuable when the server has both possible. This is particularly valuable when the server has both
Initial and Handshake data in-flight or the client has both Handshake Initial and Handshake data in flight or when the client has both
and Application Data in-flight, because the peer might only have Handshake and Application Data in flight because the peer might only
receive keys for one of the two packet number spaces. have receive keys for one of the two packet number spaces.
If the sender wants to elicit a faster acknowledgment on PTO, it can If the sender wants to elicit a faster acknowledgment on PTO, it can
skip a packet number to eliminate the acknowledgment delay. skip a packet number to eliminate the acknowledgment delay.
An endpoint SHOULD include new data in packets that are sent on PTO An endpoint SHOULD include new data in packets that are sent on PTO
expiration. Previously sent data MAY be sent if no new data can be expiration. Previously sent data MAY be sent if no new data can be
sent. Implementations MAY use alternative strategies for determining sent. Implementations MAY use alternative strategies for determining
the content of probe packets, including sending new or retransmitted the content of probe packets, including sending new or retransmitted
data based on the application's priorities. data based on the application's priorities.
It is possible the sender has no new or previously-sent data to send. It is possible the sender has no new or previously sent data to send.
As an example, consider the following sequence of events: new As an example, consider the following sequence of events: new
application data is sent in a STREAM frame, deemed lost, then application data is sent in a STREAM frame, deemed lost, then
retransmitted in a new packet, and then the original transmission is retransmitted in a new packet, and then the original transmission is
acknowledged. When there is no data to send, the sender SHOULD send acknowledged. When there is no data to send, the sender SHOULD send
a PING or other ack-eliciting frame in a single packet, re-arming the a PING or other ack-eliciting frame in a single packet, rearming the
PTO timer. PTO timer.
Alternatively, instead of sending an ack-eliciting packet, the sender Alternatively, instead of sending an ack-eliciting packet, the sender
MAY mark any packets still in flight as lost. Doing so avoids MAY mark any packets still in flight as lost. Doing so avoids
sending an additional packet, but increases the risk that loss is sending an additional packet but increases the risk that loss is
declared too aggressively, resulting in an unnecessary rate reduction declared too aggressively, resulting in an unnecessary rate reduction
by the congestion controller. by the congestion controller.
Consecutive PTO periods increase exponentially, and as a result, Consecutive PTO periods increase exponentially, and as a result,
connection recovery latency increases exponentially as packets connection recovery latency increases exponentially as packets
continue to be dropped in the network. Sending two packets on PTO continue to be dropped in the network. Sending two packets on PTO
expiration increases resilience to packet drops, thus reducing the expiration increases resilience to packet drops, thus reducing the
probability of consecutive PTO events. probability of consecutive PTO events.
When the PTO timer expires multiple times and new data cannot be When the PTO timer expires multiple times and new data cannot be
sent, implementations must choose between sending the same payload sent, implementations must choose between sending the same payload
every time or sending different payloads. Sending the same payload every time or sending different payloads. Sending the same payload
may be simpler and ensures the highest priority frames arrive first. may be simpler and ensures the highest priority frames arrive first.
Sending different payloads each time reduces the chances of spurious Sending different payloads each time reduces the chances of spurious
retransmission. retransmission.
6.3. Handling Retry Packets 6.3. Handling Retry Packets
A Retry packet causes a client to send another Initial packet, A Retry packet causes a client to send another Initial packet,
effectively restarting the connection process. A Retry packet effectively restarting the connection process. A Retry packet
indicates that the Initial was received, but not processed. A Retry indicates that the Initial packet was received but not processed. A
packet cannot be treated as an acknowledgment, because it does not Retry packet cannot be treated as an acknowledgment because it does
indicate that a packet was processed or specify the packet number. not indicate that a packet was processed or specify the packet
number.
Clients that receive a Retry packet reset congestion control and loss Clients that receive a Retry packet reset congestion control and loss
recovery state, including resetting any pending timers. Other recovery state, including resetting any pending timers. Other
connection state, in particular cryptographic handshake messages, is connection state, in particular cryptographic handshake messages, is
retained; see Section 17.2.5 of [QUIC-TRANSPORT]. retained; see Section 17.2.5 of [QUIC-TRANSPORT].
The client MAY compute an RTT estimate to the server as the time The client MAY compute an RTT estimate to the server as the time
period from when the first Initial was sent to when a Retry or a period from when the first Initial packet was sent to when a Retry or
Version Negotiation packet is received. The client MAY use this a Version Negotiation packet is received. The client MAY use this
value in place of its default for the initial RTT estimate. value in place of its default for the initial RTT estimate.
6.4. Discarding Keys and Packet State 6.4. Discarding Keys and Packet State
When Initial and Handshake packet protection keys are discarded (see When Initial and Handshake packet protection keys are discarded (see
Section 4.9 of [QUIC-TLS]), all packets that were sent with those Section 4.9 of [QUIC-TLS]), all packets that were sent with those
keys can no longer be acknowledged because their acknowledgments keys can no longer be acknowledged because their acknowledgments
cannot be processed. The sender MUST discard all recovery state cannot be processed. The sender MUST discard all recovery state
associated with those packets and MUST remove them from the count of associated with those packets and MUST remove them from the count of
bytes in flight. bytes in flight.
skipping to change at page 20, line 5 skipping to change at line 859
[QUIC-TRANSPORT]. At this point, recovery state for all in-flight [QUIC-TRANSPORT]. At this point, recovery state for all in-flight
Initial packets is discarded. Initial packets is discarded.
When 0-RTT is rejected, recovery state for all in-flight 0-RTT When 0-RTT is rejected, recovery state for all in-flight 0-RTT
packets is discarded. packets is discarded.
If a server accepts 0-RTT, but does not buffer 0-RTT packets that If a server accepts 0-RTT, but does not buffer 0-RTT packets that
arrive before Initial packets, early 0-RTT packets will be declared arrive before Initial packets, early 0-RTT packets will be declared
lost, but that is expected to be infrequent. lost, but that is expected to be infrequent.
It is expected that keys are discarded after packets encrypted with It is expected that keys are discarded at some time after the packets
them would be acknowledged or declared lost. However, Initial and encrypted with them are either acknowledged or declared lost.
Handshake secrets are discarded as soon as handshake and 1-RTT keys However, Initial and Handshake secrets are discarded as soon as
are proven to be available to both client and server; see Handshake and 1-RTT keys are proven to be available to both client
Section 4.9.1 of [QUIC-TLS]. and server; see Section 4.9.1 of [QUIC-TLS].
7. Congestion Control 7. Congestion Control
This document specifies a sender-side congestion controller for QUIC This document specifies a sender-side congestion controller for QUIC
similar to TCP NewReno ([RFC6582]). similar to TCP NewReno [RFC6582].
The signals QUIC provides for congestion control are generic and are The signals QUIC provides for congestion control are generic and are
designed to support different sender-side algorithms. A sender can designed to support different sender-side algorithms. A sender can
unilaterally choose a different algorithm to use, such as Cubic unilaterally choose a different algorithm to use, such as CUBIC
([RFC8312]). [RFC8312].
If a sender uses a different controller than that specified in this If a sender uses a different controller than that specified in this
document, the chosen controller MUST conform to the congestion document, the chosen controller MUST conform to the congestion
control guidelines specified in Section 3.1 of [RFC8085]. control guidelines specified in Section 3.1 of [RFC8085].
Similar to TCP, packets containing only ACK frames do not count Similar to TCP, packets containing only ACK frames do not count
towards bytes in flight and are not congestion controlled. Unlike toward bytes in flight and are not congestion controlled. Unlike
TCP, QUIC can detect the loss of these packets and MAY use that TCP, QUIC can detect the loss of these packets and MAY use that
information to adjust the congestion controller or the rate of ACK- information to adjust the congestion controller or the rate of ACK-
only packets being sent, but this document does not describe a only packets being sent, but this document does not describe a
mechanism for doing so. mechanism for doing so.
The congestion controller is per path, so packets sent on other paths
do not alter the current path's congestion controller, as described
in Section 9.4 of [QUIC-TRANSPORT].
The algorithm in this document specifies and uses the controller's The algorithm in this document specifies and uses the controller's
congestion window in bytes. congestion window in bytes.
An endpoint MUST NOT send a packet if it would cause bytes_in_flight An endpoint MUST NOT send a packet if it would cause bytes_in_flight
(see Appendix B.2) to be larger than the congestion window, unless (see Appendix B.2) to be larger than the congestion window, unless
the packet is sent on a PTO timer expiration (see Section 6.2) or the packet is sent on a PTO timer expiration (see Section 6.2) or
when entering recovery (see Section 7.3.2). when entering recovery (see Section 7.3.2).
7.1. Explicit Congestion Notification 7.1. Explicit Congestion Notification
If a path has been validated to support ECN ([RFC3168], [RFC8311]), If a path has been validated to support Explicit Congestion
QUIC treats a Congestion Experienced (CE) codepoint in the IP header Notification (ECN) [RFC3168] [RFC8311], QUIC treats a Congestion
as a signal of congestion. This document specifies an endpoint's Experienced (CE) codepoint in the IP header as a signal of
response when the peer-reported ECN-CE count increases; see congestion. This document specifies an endpoint's response when the
Section 13.4.2 of [QUIC-TRANSPORT]. peer-reported ECN-CE count increases; see Section 13.4.2 of
[QUIC-TRANSPORT].
7.2. Initial and Minimum Congestion Window 7.2. Initial and Minimum Congestion Window
QUIC begins every connection in slow start with the congestion window QUIC begins every connection in slow start with the congestion window
set to an initial value. Endpoints SHOULD use an initial congestion set to an initial value. Endpoints SHOULD use an initial congestion
window of 10 times the maximum datagram size (max_datagram_size), window of ten times the maximum datagram size (max_datagram_size),
while limiting the window to the larger of 14720 bytes or twice the while limiting the window to the larger of 14,720 bytes or twice the
maximum datagram size. This follows the analysis and recommendations maximum datagram size. This follows the analysis and recommendations
in [RFC6928], increasing the byte limit to account for the smaller in [RFC6928], increasing the byte limit to account for the smaller
8-byte overhead of UDP compared to the 20-byte overhead for TCP. 8-byte overhead of UDP compared to the 20-byte overhead for TCP.
If the maximum datagram size changes during the connection, the If the maximum datagram size changes during the connection, the
initial congestion window SHOULD be recalculated with the new size. initial congestion window SHOULD be recalculated with the new size.
If the maximum datagram size is decreased in order to complete the If the maximum datagram size is decreased in order to complete the
handshake, the congestion window SHOULD be set to the new initial handshake, the congestion window SHOULD be set to the new initial
congestion window. congestion window.
Prior to validating the client's address, the server can be further Prior to validating the client's address, the server can be further
limited by the anti-amplification limit as specified in Section 8.1 limited by the anti-amplification limit as specified in Section 8.1
of [QUIC-TRANSPORT]. Though the anti-amplification limit can prevent of [QUIC-TRANSPORT]. Though the anti-amplification limit can prevent
the congestion window from being fully utilized and therefore slow the congestion window from being fully utilized and therefore slow
down the increase in congestion window, it does not directly affect down the increase in congestion window, it does not directly affect
the congestion window. the congestion window.
The minimum congestion window is the smallest value the congestion The minimum congestion window is the smallest value the congestion
window can decrease to as a response to loss, increase in the peer- window can attain in response to loss, an increase in the peer-
reported ECN-CE count, or persistent congestion. The RECOMMENDED reported ECN-CE count, or persistent congestion. The RECOMMENDED
value is 2 * max_datagram_size. value is 2 * max_datagram_size.
7.3. Congestion Control States 7.3. Congestion Control States
The NewReno congestion controller described in this document has The NewReno congestion controller described in this document has
three distinct states, as shown in Figure 1. three distinct states, as shown in Figure 1.
New Path or +------------+ New path or +------------+
persistent congestion | Slow | persistent congestion | Slow |
(O)---------------------->| Start | (O)---------------------->| Start |
+------------+ +------------+
| |
Loss or | Loss or |
ECN-CE increase | ECN-CE increase |
v v
+------------+ Loss or +------------+ +------------+ Loss or +------------+
| Congestion | ECN-CE increase | Recovery | | Congestion | ECN-CE increase | Recovery |
| Avoidance |------------------>| Period | | Avoidance |------------------>| Period |
skipping to change at page 22, line 43 skipping to change at line 978
While a sender is in slow start, the congestion window increases by While a sender is in slow start, the congestion window increases by
the number of bytes acknowledged when each acknowledgment is the number of bytes acknowledged when each acknowledgment is
processed. This results in exponential growth of the congestion processed. This results in exponential growth of the congestion
window. window.
The sender MUST exit slow start and enter a recovery period when a The sender MUST exit slow start and enter a recovery period when a
packet is lost or when the ECN-CE count reported by its peer packet is lost or when the ECN-CE count reported by its peer
increases. increases.
A sender re-enters slow start any time the congestion window is less A sender reenters slow start any time the congestion window is less
than the slow start threshold, which only occurs after persistent than the slow start threshold, which only occurs after persistent
congestion is declared. congestion is declared.
7.3.2. Recovery 7.3.2. Recovery
A NewReno sender enters a recovery period when it detects the loss of A NewReno sender enters a recovery period when it detects the loss of
a packet or the ECN-CE count reported by its peer increases. A a packet or when the ECN-CE count reported by its peer increases. A
sender that is already in a recovery period stays in it and does not sender that is already in a recovery period stays in it and does not
re-enter it. reenter it.
On entering a recovery period, a sender MUST set the slow start On entering a recovery period, a sender MUST set the slow start
threshold to half the value of the congestion window when loss is threshold to half the value of the congestion window when loss is
detected. The congestion window MUST be set to the reduced value of detected. The congestion window MUST be set to the reduced value of
the slow start threshold before exiting the recovery period. the slow start threshold before exiting the recovery period.
Implementations MAY reduce the congestion window immediately upon Implementations MAY reduce the congestion window immediately upon
entering a recovery period or use other mechanisms, such as entering a recovery period or use other mechanisms, such as
Proportional Rate Reduction ([PRR]), to reduce the congestion window Proportional Rate Reduction [PRR], to reduce the congestion window
more gradually. If the congestion window is reduced immediately, a more gradually. If the congestion window is reduced immediately, a
single packet can be sent prior to reduction. This speeds up loss single packet can be sent prior to reduction. This speeds up loss
recovery if the data in the lost packet is retransmitted and is recovery if the data in the lost packet is retransmitted and is
similar to TCP as described in Section 5 of [RFC6675]. similar to TCP as described in Section 5 of [RFC6675].
The recovery period aims to limit congestion window reduction to once The recovery period aims to limit congestion window reduction to once
per round trip. Therefore during a recovery period, the congestion per round trip. Therefore, during a recovery period, the congestion
window does not change in response to new losses or increases in the window does not change in response to new losses or increases in the
ECN-CE count. ECN-CE count.
A recovery period ends and the sender enters congestion avoidance A recovery period ends and the sender enters congestion avoidance
when a packet sent during the recovery period is acknowledged. This when a packet sent during the recovery period is acknowledged. This
is slightly different from TCP's definition of recovery, which ends is slightly different from TCP's definition of recovery, which ends
when the lost segment that started recovery is acknowledged when the lost segment that started recovery is acknowledged
([RFC5681]). [RFC5681].
7.3.3. Congestion Avoidance 7.3.3. Congestion Avoidance
A NewReno sender is in congestion avoidance any time the congestion A NewReno sender is in congestion avoidance any time the congestion
window is at or above the slow start threshold and not in a recovery window is at or above the slow start threshold and not in a recovery
period. period.
A sender in congestion avoidance uses an Additive Increase A sender in congestion avoidance uses an Additive Increase
Multiplicative Decrease (AIMD) approach that MUST limit the increase Multiplicative Decrease (AIMD) approach that MUST limit the increase
to the congestion window to at most one maximum datagram size for to the congestion window to at most one maximum datagram size for
each congestion window that is acknowledged. each congestion window that is acknowledged.
The sender exits congestion avoidance and enters a recovery period The sender exits congestion avoidance and enters a recovery period
when a packet is lost or when the ECN-CE count reported by its peer when a packet is lost or when the ECN-CE count reported by its peer
increases. increases.
7.4. Ignoring Loss of Undecryptable Packets 7.4. Ignoring Loss of Undecryptable Packets
During the handshake, some packet protection keys might not be During the handshake, some packet protection keys might not be
available when a packet arrives and the receiver can choose to drop available when a packet arrives, and the receiver can choose to drop
the packet. In particular, Handshake and 0-RTT packets cannot be the packet. In particular, Handshake and 0-RTT packets cannot be
processed until the Initial packets arrive and 1-RTT packets cannot processed until the Initial packets arrive, and 1-RTT packets cannot
be processed until the handshake completes. Endpoints MAY ignore the be processed until the handshake completes. Endpoints MAY ignore the
loss of Handshake, 0-RTT, and 1-RTT packets that might have arrived loss of Handshake, 0-RTT, and 1-RTT packets that might have arrived
before the peer had packet protection keys to process those packets. before the peer had packet protection keys to process those packets.
Endpoints MUST NOT ignore the loss of packets that were sent after Endpoints MUST NOT ignore the loss of packets that were sent after
the earliest acknowledged packet in a given packet number space. the earliest acknowledged packet in a given packet number space.
7.5. Probe Timeout 7.5. Probe Timeout
Probe packets MUST NOT be blocked by the congestion controller. A Probe packets MUST NOT be blocked by the congestion controller. A
sender MUST however count these packets as being additionally in sender MUST however count these packets as being additionally in
flight, since these packets add network load without establishing flight, since these packets add network load without establishing
packet loss. Note that sending probe packets might cause the packet loss. Note that sending probe packets might cause the
sender's bytes in flight to exceed the congestion window until an sender's bytes in flight to exceed the congestion window until an
skipping to change at page 24, line 37 skipping to change at line 1069
(smoothed_rtt + max(4*rttvar, kGranularity) + max_ack_delay) * (smoothed_rtt + max(4*rttvar, kGranularity) + max_ack_delay) *
kPersistentCongestionThreshold kPersistentCongestionThreshold
Unlike the PTO computation in Section 6.2, this duration includes the Unlike the PTO computation in Section 6.2, this duration includes the
max_ack_delay irrespective of the packet number spaces in which max_ack_delay irrespective of the packet number spaces in which
losses are established. losses are established.
This duration allows a sender to send as many packets before This duration allows a sender to send as many packets before
establishing persistent congestion, including some in response to PTO establishing persistent congestion, including some in response to PTO
expiration, as TCP does with Tail Loss Probes ([RACK]) and a expiration, as TCP does with Tail Loss Probes [RFC8985] and an RTO
Retransmission Timeout ([RFC5681]). [RFC5681].
Larger values of kPersistentCongestionThreshold cause the sender to Larger values of kPersistentCongestionThreshold cause the sender to
become less responsive to persistent congestion in the network, which become less responsive to persistent congestion in the network, which
can result in aggressive sending into a congested network. Too small can result in aggressive sending into a congested network. Too small
a value can result in a sender declaring persistent congestion a value can result in a sender declaring persistent congestion
unnecessarily, resulting in reduced throughput for the sender. unnecessarily, resulting in reduced throughput for the sender.
The RECOMMENDED value for kPersistentCongestionThreshold is 3, which The RECOMMENDED value for kPersistentCongestionThreshold is 3, which
results in behavior that is approximately equivalent to a TCP sender results in behavior that is approximately equivalent to a TCP sender
declaring an RTO after two TLPs. declaring an RTO after two TLPs.
This design does not use consecutive PTO events to establish This design does not use consecutive PTO events to establish
persistent congestion, since application patterns impact PTO persistent congestion, since application patterns impact PTO
expirations. For example, a sender that sends small amounts of data expiration. For example, a sender that sends small amounts of data
with silence periods between them restarts the PTO timer every time with silence periods between them restarts the PTO timer every time
it sends, potentially preventing the PTO timer from expiring for a it sends, potentially preventing the PTO timer from expiring for a
long period of time, even when no acknowledgments are being received. long period of time, even when no acknowledgments are being received.
The use of a duration enables a sender to establish persistent The use of a duration enables a sender to establish persistent
congestion without depending on PTO expiration. congestion without depending on PTO expiration.
7.6.2. Establishing Persistent Congestion 7.6.2. Establishing Persistent Congestion
A sender establishes persistent congestion after the receipt of an A sender establishes persistent congestion after the receipt of an
acknowledgment if two packets that are ack-eliciting are declared acknowledgment if two packets that are ack-eliciting are declared
skipping to change at page 25, line 25 skipping to change at line 1106
* across all packet number spaces, none of the packets sent between * across all packet number spaces, none of the packets sent between
the send times of these two packets are acknowledged; the send times of these two packets are acknowledged;
* the duration between the send times of these two packets exceeds * the duration between the send times of these two packets exceeds
the persistent congestion duration (Section 7.6.1); and the persistent congestion duration (Section 7.6.1); and
* a prior RTT sample existed when these two packets were sent. * a prior RTT sample existed when these two packets were sent.
These two packets MUST be ack-eliciting, since a receiver is required These two packets MUST be ack-eliciting, since a receiver is required
to acknowledge only ack-eliciting packets within its maximum ack to acknowledge only ack-eliciting packets within its maximum
delay; see Section 13.2 of [QUIC-TRANSPORT]. acknowledgment delay; see Section 13.2 of [QUIC-TRANSPORT].
The persistent congestion period SHOULD NOT start until there is at The persistent congestion period SHOULD NOT start until there is at
least one RTT sample. Before the first RTT sample, a sender arms its least one RTT sample. Before the first RTT sample, a sender arms its
PTO timer based on the initial RTT (Section 6.2.2), which could be PTO timer based on the initial RTT (Section 6.2.2), which could be
substantially larger than the actual RTT. Requiring a prior RTT substantially larger than the actual RTT. Requiring a prior RTT
sample prevents a sender from establishing persistent congestion with sample prevents a sender from establishing persistent congestion with
potentially too few probes. potentially too few probes.
Since network congestion is not affected by packet number spaces, Since network congestion is not affected by packet number spaces,
persistent congestion SHOULD consider packets sent across packet persistent congestion SHOULD consider packets sent across packet
number spaces. A sender that does not have state for all packet number spaces. A sender that does not have state for all packet
number spaces or an implementation that cannot compare send times number spaces or an implementation that cannot compare send times
across packet number spaces MAY use state for just the packet number across packet number spaces MAY use state for just the packet number
space that was acknowledged. This might result in erroneously space that was acknowledged. This might result in erroneously
declaring persistent congestion, but it will not lead to a failure to declaring persistent congestion, but it will not lead to a failure to
detect persistent congestion. detect persistent congestion.
When persistent congestion is declared, the sender's congestion When persistent congestion is declared, the sender's congestion
window MUST be reduced to the minimum congestion window window MUST be reduced to the minimum congestion window
(kMinimumWindow), similar to a TCP sender's response on an RTO (kMinimumWindow), similar to a TCP sender's response on an RTO
([RFC5681]). [RFC5681].
7.6.3. Example 7.6.3. Example
The following example illustrates how a sender might establish The following example illustrates how a sender might establish
persistent congestion. Assume: persistent congestion. Assume:
smoothed_rtt + max(4*rttvar, kGranularity) + max_ack_delay = 2 smoothed_rtt + max(4*rttvar, kGranularity) + max_ack_delay = 2
kPersistentCongestionThreshold = 3 kPersistentCongestionThreshold = 3
Consider the following sequence of events: Consider the following sequence of events:
+========+===========================+ +========+===================================+
| Time | Action | | Time | Action |
+========+===========================+ +========+===================================+
| t=0 | Send packet #1 (app data) | | t=0 | Send packet #1 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=1 | Send packet #2 (app data) | | t=1 | Send packet #2 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=1.2 | Recv acknowledgment of #1 | | t=1.2 | Receive acknowledgment of #1 |
+--------+---------------------------+ +--------+-----------------------------------+
| t=2 | Send packet #3 (app data) | | t=2 | Send packet #3 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=3 | Send packet #4 (app data) | | t=3 | Send packet #4 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=4 | Send packet #5 (app data) | | t=4 | Send packet #5 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=5 | Send packet #6 (app data) | | t=5 | Send packet #6 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=6 | Send packet #7 (app data) | | t=6 | Send packet #7 (application data) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=8 | Send packet #8 (PTO 1) | | t=8 | Send packet #8 (PTO 1) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=12 | Send packet #9 (PTO 2) | | t=12 | Send packet #9 (PTO 2) |
+--------+---------------------------+ +--------+-----------------------------------+
| t=12.2 | Recv acknowledgment of #9 | | t=12.2 | Receive acknowledgment of #9 |
+--------+---------------------------+ +--------+-----------------------------------+
Table 1 Table 1
Packets 2 through 8 are declared lost when the acknowledgment for Packets 2 through 8 are declared lost when the acknowledgment for
packet 9 is received at t = 12.2. packet 9 is received at "t = 12.2".
The congestion period is calculated as the time between the oldest The congestion period is calculated as the time between the oldest
and newest lost packets: 8 - 1 = 7. The persistent congestion and newest lost packets: "8 - 1 = 7". The persistent congestion
duration is: 2 * 3 = 6. Because the threshold was reached and duration is "2 * 3 = 6". Because the threshold was reached and
because none of the packets between the oldest and the newest lost because none of the packets between the oldest and the newest lost
packets were acknowledged, the network is considered to have packets were acknowledged, the network is considered to have
experienced persistent congestion. experienced persistent congestion.
While this example shows PTO expiration, they are not required for While this example shows PTO expiration, they are not required for
persistent congestion to be established. persistent congestion to be established.
7.7. Pacing 7.7. Pacing
A sender SHOULD pace sending of all in-flight packets based on input A sender SHOULD pace sending of all in-flight packets based on input
skipping to change at page 27, line 27 skipping to change at line 1200
Section 7.2. A sender with knowledge that the network path to the Section 7.2. A sender with knowledge that the network path to the
receiver can absorb larger bursts MAY use a higher limit. receiver can absorb larger bursts MAY use a higher limit.
An implementation should take care to architect its congestion An implementation should take care to architect its congestion
controller to work well with a pacer. For instance, a pacer might controller to work well with a pacer. For instance, a pacer might
wrap the congestion controller and control the availability of the wrap the congestion controller and control the availability of the
congestion window, or a pacer might pace out packets handed to it by congestion window, or a pacer might pace out packets handed to it by
the congestion controller. the congestion controller.
Timely delivery of ACK frames is important for efficient loss Timely delivery of ACK frames is important for efficient loss
recovery. Packets containing only ACK frames SHOULD therefore not be recovery. To avoid delaying their delivery to the peer, packets
paced, to avoid delaying their delivery to the peer. containing only ACK frames SHOULD therefore not be paced.
Endpoints can implement pacing as they choose. A perfectly paced Endpoints can implement pacing as they choose. A perfectly paced
sender spreads packets exactly evenly over time. For a window-based sender spreads packets exactly evenly over time. For a window-based
congestion controller, such as the one in this document, that rate congestion controller, such as the one in this document, that rate
can be computed by averaging the congestion window over the round- can be computed by averaging the congestion window over the RTT.
trip time. Expressed as a rate in units of bytes per time, where Expressed as a rate in units of bytes per time, where
congestion_window is in bytes: congestion_window is in bytes:
rate = N * congestion_window / smoothed_rtt rate = N * congestion_window / smoothed_rtt
Or, expressed as an inter-packet interval in units of time: Or expressed as an inter-packet interval in units of time:
interval = ( smoothed_rtt * packet_size / congestion_window ) / N interval = ( smoothed_rtt * packet_size / congestion_window ) / N
Using a value for "N" that is small, but at least 1 (for example, Using a value for "N" that is small, but at least 1 (for example,
1.25) ensures that variations in round-trip time do not result in 1.25) ensures that variations in RTT do not result in
under-utilization of the congestion window. underutilization of the congestion window.
Practical considerations, such as packetization, scheduling delays, Practical considerations, such as packetization, scheduling delays,
and computational efficiency, can cause a sender to deviate from this and computational efficiency, can cause a sender to deviate from this
rate over time periods that are much shorter than a round-trip time. rate over time periods that are much shorter than an RTT.
One possible implementation strategy for pacing uses a leaky bucket One possible implementation strategy for pacing uses a leaky bucket
algorithm, where the capacity of the "bucket" is limited to the algorithm, where the capacity of the "bucket" is limited to the
maximum burst size and the rate the "bucket" fills is determined by maximum burst size and the rate the "bucket" fills is determined by
the above function. the above function.
7.8. Under-utilizing the Congestion Window 7.8. Underutilizing the Congestion Window
When bytes in flight is smaller than the congestion window and When bytes in flight is smaller than the congestion window and
sending is not pacing limited, the congestion window is under- sending is not pacing limited, the congestion window is
utilized. When this occurs, the congestion window SHOULD NOT be underutilized. This can happen due to insufficient application data
increased in either slow start or congestion avoidance. This can or flow control limits. When this occurs, the congestion window
happen due to insufficient application data or flow control limits. SHOULD NOT be increased in either slow start or congestion avoidance.
A sender that paces packets (see Section 7.7) might delay sending A sender that paces packets (see Section 7.7) might delay sending
packets and not fully utilize the congestion window due to this packets and not fully utilize the congestion window due to this
delay. A sender SHOULD NOT consider itself application limited if it delay. A sender SHOULD NOT consider itself application limited if it
would have fully utilized the congestion window without pacing delay. would have fully utilized the congestion window without pacing delay.
A sender MAY implement alternative mechanisms to update its A sender MAY implement alternative mechanisms to update its
congestion window after periods of under-utilization, such as those congestion window after periods of underutilization, such as those
proposed for TCP in [RFC7661]. proposed for TCP in [RFC7661].
8. Security Considerations 8. Security Considerations
8.1. Loss and Congestion Signals 8.1. Loss and Congestion Signals
Loss detection and congestion control fundamentally involve Loss detection and congestion control fundamentally involve the
consumption of signals, such as delay, loss, and ECN markings, from consumption of signals, such as delay, loss, and ECN markings, from
unauthenticated entities. An attacker can cause endpoints to reduce unauthenticated entities. An attacker can cause endpoints to reduce
their sending rate by manipulating these signals; by dropping their sending rate by manipulating these signals: by dropping
packets, by altering path delay strategically, or by changing ECN packets, by altering path delay strategically, or by changing ECN
codepoints. codepoints.
8.2. Traffic Analysis 8.2. Traffic Analysis
Packets that carry only ACK frames can be heuristically identified by Packets that carry only ACK frames can be heuristically identified by
observing packet size. Acknowledgment patterns may expose observing packet size. Acknowledgment patterns may expose
information about link characteristics or application behavior. To information about link characteristics or application behavior. To
reduce leaked information, endpoints can bundle acknowledgments with reduce leaked information, endpoints can bundle acknowledgments with
other frames, or they can use PADDING frames at a potential cost to other frames, or they can use PADDING frames at a potential cost to
skipping to change at page 29, line 9 skipping to change at line 1277
A receiver can misreport ECN markings to alter the congestion A receiver can misreport ECN markings to alter the congestion
response of a sender. Suppressing reports of ECN-CE markings could response of a sender. Suppressing reports of ECN-CE markings could
cause a sender to increase their send rate. This increase could cause a sender to increase their send rate. This increase could
result in congestion and loss. result in congestion and loss.
A sender can detect suppression of reports by marking occasional A sender can detect suppression of reports by marking occasional
packets that it sends with an ECN-CE marking. If a packet sent with packets that it sends with an ECN-CE marking. If a packet sent with
an ECN-CE marking is not reported as having been CE marked when the an ECN-CE marking is not reported as having been CE marked when the
packet is acknowledged, then the sender can disable ECN for that path packet is acknowledged, then the sender can disable ECN for that path
by not setting ECT codepoints in subsequent packets sent on that path by not setting ECN-Capable Transport (ECT) codepoints in subsequent
[RFC3168]. packets sent on that path [RFC3168].
Reporting additional ECN-CE markings will cause a sender to reduce Reporting additional ECN-CE markings will cause a sender to reduce
their sending rate, which is similar in effect to advertising reduced their sending rate, which is similar in effect to advertising reduced
connection flow control limits and so no advantage is gained by doing connection flow control limits and so no advantage is gained by doing
so. so.
Endpoints choose the congestion controller that they use. Congestion Endpoints choose the congestion controller that they use. Congestion
controllers respond to reports of ECN-CE by reducing their rate, but controllers respond to reports of ECN-CE by reducing their rate, but
the response may vary. Markings can be treated as equivalent to loss the response may vary. Markings can be treated as equivalent to loss
([RFC3168]), but other responses can be specified, such as [RFC3168], but other responses can be specified, such as [RFC8511] or
([RFC8511]) or ([RFC8311]). [RFC8311].
9. IANA Considerations
This document has no IANA actions.
10. References 9. References
10.1. Normative References 9.1. Normative References
[QUIC-TLS] Thomson, M., Ed. and S. Turner, Ed., "Using TLS to Secure [QUIC-TLS] Thomson, M., Ed. and S. Turner, Ed., "Using TLS to Secure
QUIC", Work in Progress, Internet-Draft, draft-ietf-quic- QUIC", RFC 9001, DOI 10.17487/RFC9001, May 2021,
tls-34, 15 January 2021, <https://www.rfc-editor.org/info/rfc9001>.
<https://tools.ietf.org/html/draft-ietf-quic-tls-34>.
[QUIC-TRANSPORT] [QUIC-TRANSPORT]
Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based
Multiplexed and Secure Transport", Work in Progress, Multiplexed and Secure Transport", RFC 9000,
Internet-Draft, draft-ietf-quic-transport-34, 15 January DOI 10.17487/RFC9000, May 2021,
2021, <https://tools.ietf.org/html/draft-ietf-quic- <https://www.rfc-editor.org/info/rfc9000>.
transport-34>.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997, DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>. <https://www.rfc-editor.org/info/rfc2119>.
[RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition
of Explicit Congestion Notification (ECN) to IP", of Explicit Congestion Notification (ECN) to IP",
RFC 3168, DOI 10.17487/RFC3168, September 2001, RFC 3168, DOI 10.17487/RFC3168, September 2001,
<https://www.rfc-editor.org/info/rfc3168>. <https://www.rfc-editor.org/info/rfc3168>.
[RFC8085] Eggert, L., Fairhurst, G., and G. Shepherd, "UDP Usage [RFC8085] Eggert, L., Fairhurst, G., and G. Shepherd, "UDP Usage
Guidelines", BCP 145, RFC 8085, DOI 10.17487/RFC8085, Guidelines", BCP 145, RFC 8085, DOI 10.17487/RFC8085,
March 2017, <https://www.rfc-editor.org/info/rfc8085>. March 2017, <https://www.rfc-editor.org/info/rfc8085>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
May 2017, <https://www.rfc-editor.org/info/rfc8174>. May 2017, <https://www.rfc-editor.org/info/rfc8174>.
10.2. Informative References 9.2. Informative References
[FACK] Mathis, M. and J. Mahdavi, "Forward Acknowledgement: [FACK] Mathis, M. and J. Mahdavi, "Forward acknowledgement:
Refining TCP Congestion Control", ACM SIGCOMM , August Refining TCP Congestion Control", ACM SIGCOMM Computer
1996. Communication Review, DOI 10.1145/248157.248181, August
1996, <https://doi.org/10.1145/248157.248181>.
[PRR] Mathis, M., Dukkipati, N., and Y. Cheng, "Proportional [PRR] Mathis, M., Dukkipati, N., and Y. Cheng, "Proportional
Rate Reduction for TCP", RFC 6937, DOI 10.17487/RFC6937, Rate Reduction for TCP", RFC 6937, DOI 10.17487/RFC6937,
May 2013, <https://www.rfc-editor.org/info/rfc6937>. May 2013, <https://www.rfc-editor.org/info/rfc6937>.
[RACK] Cheng, Y., Cardwell, N., Dukkipati, N., and P. Jha, "The
RACK-TLP loss detection algorithm for TCP", Work in
Progress, Internet-Draft, draft-ietf-tcpm-rack-15, 22
December 2020, <http://www.ietf.org/internet-drafts/draft-
ietf-tcpm-rack-15.txt>.
[RETRANSMISSION] [RETRANSMISSION]
Karn, P. and C. Partridge, "Improving Round-Trip Time Karn, P. and C. Partridge, "Improving Round-Trip Time
Estimates in Reliable Transport Protocols", ACM SIGCOMM Estimates in Reliable Transport Protocols", ACM
CCR , January 1995. Transactions on Computer Systems,
DOI 10.1145/118544.118549, November 1991,
<https://doi.org/10.1145/118544.118549>.
[RFC2018] Mathis, M., Mahdavi, J., Floyd, S., and A. Romanow, "TCP [RFC2018] Mathis, M., Mahdavi, J., Floyd, S., and A. Romanow, "TCP
Selective Acknowledgment Options", RFC 2018, Selective Acknowledgment Options", RFC 2018,
DOI 10.17487/RFC2018, October 1996, DOI 10.17487/RFC2018, October 1996,
<https://www.rfc-editor.org/info/rfc2018>. <https://www.rfc-editor.org/info/rfc2018>.
[RFC3465] Allman, M., "TCP Congestion Control with Appropriate Byte [RFC3465] Allman, M., "TCP Congestion Control with Appropriate Byte
Counting (ABC)", RFC 3465, DOI 10.17487/RFC3465, February Counting (ABC)", RFC 3465, DOI 10.17487/RFC3465, February
2003, <https://www.rfc-editor.org/info/rfc3465>. 2003, <https://www.rfc-editor.org/info/rfc3465>.
skipping to change at page 31, line 11 skipping to change at line 1366
Spurious Retransmission Timeouts with TCP", RFC 5682, Spurious Retransmission Timeouts with TCP", RFC 5682,
DOI 10.17487/RFC5682, September 2009, DOI 10.17487/RFC5682, September 2009,
<https://www.rfc-editor.org/info/rfc5682>. <https://www.rfc-editor.org/info/rfc5682>.
[RFC5827] Allman, M., Avrachenkov, K., Ayesta, U., Blanton, J., and [RFC5827] Allman, M., Avrachenkov, K., Ayesta, U., Blanton, J., and
P. Hurtig, "Early Retransmit for TCP and Stream Control P. Hurtig, "Early Retransmit for TCP and Stream Control
Transmission Protocol (SCTP)", RFC 5827, Transmission Protocol (SCTP)", RFC 5827,
DOI 10.17487/RFC5827, May 2010, DOI 10.17487/RFC5827, May 2010,
<https://www.rfc-editor.org/info/rfc5827>. <https://www.rfc-editor.org/info/rfc5827>.
[RFC6297] Welzl, M. and D. Ros, "A Survey of Lower-than-Best-Effort
Transport Protocols", RFC 6297, DOI 10.17487/RFC6297, June
2011, <https://www.rfc-editor.org/info/rfc6297>.
[RFC6298] Paxson, V., Allman, M., Chu, J., and M. Sargent, [RFC6298] Paxson, V., Allman, M., Chu, J., and M. Sargent,
"Computing TCP's Retransmission Timer", RFC 6298, "Computing TCP's Retransmission Timer", RFC 6298,
DOI 10.17487/RFC6298, June 2011, DOI 10.17487/RFC6298, June 2011,
<https://www.rfc-editor.org/info/rfc6298>. <https://www.rfc-editor.org/info/rfc6298>.
[RFC6582] Henderson, T., Floyd, S., Gurtov, A., and Y. Nishida, "The [RFC6582] Henderson, T., Floyd, S., Gurtov, A., and Y. Nishida, "The
NewReno Modification to TCP's Fast Recovery Algorithm", NewReno Modification to TCP's Fast Recovery Algorithm",
RFC 6582, DOI 10.17487/RFC6582, April 2012, RFC 6582, DOI 10.17487/RFC6582, April 2012,
<https://www.rfc-editor.org/info/rfc6582>. <https://www.rfc-editor.org/info/rfc6582>.
skipping to change at page 32, line 10 skipping to change at line 1407
[RFC8312] Rhee, I., Xu, L., Ha, S., Zimmermann, A., Eggert, L., and [RFC8312] Rhee, I., Xu, L., Ha, S., Zimmermann, A., Eggert, L., and
R. Scheffenegger, "CUBIC for Fast Long-Distance Networks", R. Scheffenegger, "CUBIC for Fast Long-Distance Networks",
RFC 8312, DOI 10.17487/RFC8312, February 2018, RFC 8312, DOI 10.17487/RFC8312, February 2018,
<https://www.rfc-editor.org/info/rfc8312>. <https://www.rfc-editor.org/info/rfc8312>.
[RFC8511] Khademi, N., Welzl, M., Armitage, G., and G. Fairhurst, [RFC8511] Khademi, N., Welzl, M., Armitage, G., and G. Fairhurst,
"TCP Alternative Backoff with ECN (ABE)", RFC 8511, "TCP Alternative Backoff with ECN (ABE)", RFC 8511,
DOI 10.17487/RFC8511, December 2018, DOI 10.17487/RFC8511, December 2018,
<https://www.rfc-editor.org/info/rfc8511>. <https://www.rfc-editor.org/info/rfc8511>.
[RFC8985] Cheng, Y., Cardwell, N., Dukkipati, N., and P. Jha, "The
RACK-TLP Loss Detection Algorithm for TCP", RFC 8985,
DOI 10.17487/RFC8985, February 2021,
<https://www.rfc-editor.org/info/rfc8985>.
Appendix A. Loss Recovery Pseudocode Appendix A. Loss Recovery Pseudocode
We now describe an example implementation of the loss detection We now describe an example implementation of the loss detection
mechanisms described in Section 6. mechanisms described in Section 6.
The pseudocode segments in this section are licensed as Code The pseudocode segments in this section are licensed as Code
Components; see the copyright notice. Components; see the copyright notice.
A.1. Tracking Sent Packets A.1. Tracking Sent Packets
skipping to change at page 32, line 38 skipping to change at line 1440
see Section 13.3 of [QUIC-TRANSPORT]. This enables a sender to see Section 13.3 of [QUIC-TRANSPORT]. This enables a sender to
detect spurious retransmissions. detect spurious retransmissions.
Sent packets are tracked for each packet number space, and ACK Sent packets are tracked for each packet number space, and ACK
processing only applies to a single space. processing only applies to a single space.
A.1.1. Sent Packet Fields A.1.1. Sent Packet Fields
packet_number: The packet number of the sent packet. packet_number: The packet number of the sent packet.
ack_eliciting: A boolean that indicates whether a packet is ack- ack_eliciting: A Boolean that indicates whether a packet is ack-
eliciting. If true, it is expected that an acknowledgment will be eliciting. If true, it is expected that an acknowledgment will be
received, though the peer could delay sending the ACK frame received, though the peer could delay sending the ACK frame
containing it by up to the max_ack_delay. containing it by up to the max_ack_delay.
in_flight: A boolean that indicates whether the packet counts in_flight: A Boolean that indicates whether the packet counts toward
towards bytes in flight. bytes in flight.
sent_bytes: The number of bytes sent in the packet, not including sent_bytes: The number of bytes sent in the packet, not including
UDP or IP overhead, but including QUIC framing overhead. UDP or IP overhead, but including QUIC framing overhead.
time_sent: The time the packet was sent. time_sent: The time the packet was sent.
A.2. Constants of Interest A.2. Constants of Interest
Constants used in loss recovery are based on a combination of RFCs, Constants used in loss recovery are based on a combination of RFCs,
papers, and common practice. papers, and common practice.
kPacketThreshold: Maximum reordering in packets before packet kPacketThreshold: Maximum reordering in packets before packet
threshold loss detection considers a packet lost. The value threshold loss detection considers a packet lost. The value
recommended in Section 6.1.1 is 3. recommended in Section 6.1.1 is 3.
kTimeThreshold: Maximum reordering in time before time threshold kTimeThreshold: Maximum reordering in time before time threshold
loss detection considers a packet lost. Specified as an RTT loss detection considers a packet lost. Specified as an RTT
multiplier. The value recommended in Section 6.1.2 is 9/8. multiplier. The value recommended in Section 6.1.2 is 9/8.
kGranularity: Timer granularity. This is a system-dependent value, kGranularity: Timer granularity. This is a system-dependent value,
and Section 6.1.2 recommends a value of 1ms. and Section 6.1.2 recommends a value of 1 ms.
kInitialRtt: The RTT used before an RTT sample is taken. The value kInitialRtt: The RTT used before an RTT sample is taken. The value
recommended in Section 6.2.2 is 333ms. recommended in Section 6.2.2 is 333 ms.
kPacketNumberSpace: An enum to enumerate the three packet number kPacketNumberSpace: An enum to enumerate the three packet number
spaces. spaces:
enum kPacketNumberSpace { enum kPacketNumberSpace {
Initial, Initial,
Handshake, Handshake,
ApplicationData, ApplicationData,
} }
A.3. Variables of interest A.3. Variables of Interest
Variables required to implement the congestion control mechanisms are Variables required to implement the congestion control mechanisms are
described in this section. described in this section.
latest_rtt: The most recent RTT measurement made when receiving an latest_rtt: The most recent RTT measurement made when receiving an
ack for a previously unacked packet. acknowledgment for a previously unacknowledged packet.
smoothed_rtt: The smoothed RTT of the connection, computed as smoothed_rtt: The smoothed RTT of the connection, computed as
described in Section 5.3. described in Section 5.3.
rttvar: The RTT variation, computed as described in Section 5.3. rttvar: The RTT variation, computed as described in Section 5.3.
min_rtt: The minimum RTT seen over a period of time, ignoring min_rtt: The minimum RTT seen over a period of time, ignoring
acknowledgment delay, as described in Section 5.2. acknowledgment delay, as described in Section 5.2.
first_rtt_sample: The time that the first RTT sample was obtained. first_rtt_sample: The time that the first RTT sample was obtained.
skipping to change at page 34, line 13 skipping to change at line 1509
max_ack_delay: The maximum amount of time by which the receiver max_ack_delay: The maximum amount of time by which the receiver
intends to delay acknowledgments for packets in the Application intends to delay acknowledgments for packets in the Application
Data packet number space, as defined by the eponymous transport Data packet number space, as defined by the eponymous transport
parameter (Section 18.2 of [QUIC-TRANSPORT]). Note that the parameter (Section 18.2 of [QUIC-TRANSPORT]). Note that the
actual ack_delay in a received ACK frame may be larger due to late actual ack_delay in a received ACK frame may be larger due to late
timers, reordering, or loss. timers, reordering, or loss.
loss_detection_timer: Multi-modal timer used for loss detection. loss_detection_timer: Multi-modal timer used for loss detection.
pto_count: The number of times a PTO has been sent without receiving pto_count: The number of times a PTO has been sent without receiving
an ack. an acknowledgment.
time_of_last_ack_eliciting_packet[kPacketNumberSpace]: The time the time_of_last_ack_eliciting_packet[kPacketNumberSpace]: The time the
most recent ack-eliciting packet was sent. most recent ack-eliciting packet was sent.
largest_acked_packet[kPacketNumberSpace]: The largest packet number largest_acked_packet[kPacketNumberSpace]: The largest packet number
acknowledged in the packet number space so far. acknowledged in the packet number space so far.
loss_time[kPacketNumberSpace]: The time at which the next packet in loss_time[kPacketNumberSpace]: The time at which the next packet in
that packet number space can be considered lost based on exceeding that packet number space can be considered lost based on exceeding
the reordering window in time. the reordering window in time.
skipping to change at page 35, line 27 skipping to change at line 1570
if (ack_eliciting): if (ack_eliciting):
time_of_last_ack_eliciting_packet[pn_space] = now() time_of_last_ack_eliciting_packet[pn_space] = now()
OnPacketSentCC(sent_bytes) OnPacketSentCC(sent_bytes)
SetLossDetectionTimer() SetLossDetectionTimer()
A.6. On Receiving a Datagram A.6. On Receiving a Datagram
When a server is blocked by anti-amplification limits, receiving a When a server is blocked by anti-amplification limits, receiving a
datagram unblocks it, even if none of the packets in the datagram are datagram unblocks it, even if none of the packets in the datagram are
successfully processed. In such a case, the PTO timer will need to successfully processed. In such a case, the PTO timer will need to
be re-armed. be rearmed.
Pseudocode for OnDatagramReceived follows: Pseudocode for OnDatagramReceived follows:
OnDatagramReceived(datagram): OnDatagramReceived(datagram):
// If this datagram unblocks the server, arm the // If this datagram unblocks the server, arm the
// PTO timer to avoid deadlock. // PTO timer to avoid deadlock.
if (server was at anti-amplification limit): if (server was at anti-amplification limit):
SetLossDetectionTimer() SetLossDetectionTimer()
if loss_detection_timer.timeout < now():
// Execute PTO if it would have expired
// while the amplification limit applied.
OnLossDetectionTimeout()
A.7. On Receiving an Acknowledgment A.7. On Receiving an Acknowledgment
When an ACK frame is received, it may newly acknowledge any number of When an ACK frame is received, it may newly acknowledge any number of
packets. packets.
Pseudocode for OnAckReceived and UpdateRtt follow: Pseudocode for OnAckReceived and UpdateRtt follow:
IncludesAckEliciting(packets): IncludesAckEliciting(packets):
for packet in packets: for packet in packets:
skipping to change at page 37, line 10 skipping to change at line 1653
// min_rtt ignores acknowledgment delay. // min_rtt ignores acknowledgment delay.
min_rtt = min(min_rtt, latest_rtt) min_rtt = min(min_rtt, latest_rtt)
// Limit ack_delay by max_ack_delay after handshake // Limit ack_delay by max_ack_delay after handshake
// confirmation. // confirmation.
if (handshake confirmed): if (handshake confirmed):
ack_delay = min(ack_delay, max_ack_delay) ack_delay = min(ack_delay, max_ack_delay)
// Adjust for acknowledgment delay if plausible. // Adjust for acknowledgment delay if plausible.
adjusted_rtt = latest_rtt adjusted_rtt = latest_rtt
if (latest_rtt > min_rtt + ack_delay): if (latest_rtt >= min_rtt + ack_delay):
adjusted_rtt = latest_rtt - ack_delay adjusted_rtt = latest_rtt - ack_delay
rttvar = 3/4 * rttvar + 1/4 * abs(smoothed_rtt - adjusted_rtt) rttvar = 3/4 * rttvar + 1/4 * abs(smoothed_rtt - adjusted_rtt)
smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * adjusted_rtt smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * adjusted_rtt
A.8. Setting the Loss Detection Timer A.8. Setting the Loss Detection Timer
QUIC loss detection uses a single timer for all timeout loss QUIC loss detection uses a single timer for all timeout loss
detection. The duration of the timer is based on the timer's mode, detection. The duration of the timer is based on the timer's mode,
which is set in the packet and timer events further below. The which is set in the packet and timer events further below. The
skipping to change at page 37, line 43 skipping to change at line 1686
space = Initial space = Initial
for pn_space in [ Handshake, ApplicationData ]: for pn_space in [ Handshake, ApplicationData ]:
if (time == 0 || loss_time[pn_space] < time): if (time == 0 || loss_time[pn_space] < time):
time = loss_time[pn_space]; time = loss_time[pn_space];
space = pn_space space = pn_space
return time, space return time, space
GetPtoTimeAndSpace(): GetPtoTimeAndSpace():
duration = (smoothed_rtt + max(4 * rttvar, kGranularity)) duration = (smoothed_rtt + max(4 * rttvar, kGranularity))
* (2 ^ pto_count) * (2 ^ pto_count)
// Arm PTO from now when there are no inflight packets. // Anti-deadlock PTO starts from the current time
if (no in-flight packets): if (no ack-eliciting packets in flight):
assert(!PeerCompletedAddressValidation()) assert(!PeerCompletedAddressValidation())
if (has handshake keys): if (has handshake keys):
return (now() + duration), Handshake return (now() + duration), Handshake
else: else:
return (now() + duration), Initial return (now() + duration), Initial
pto_timeout = infinite pto_timeout = infinite
pto_space = Initial pto_space = Initial
for space in [ Initial, Handshake, ApplicationData ]: for space in [ Initial, Handshake, ApplicationData ]:
if (no in-flight packets in space): if (no ack-eliciting packets in flight in space):
continue; continue;
if (space == ApplicationData): if (space == ApplicationData):
// Skip Application Data until handshake confirmed. // Skip Application Data until handshake confirmed.
if (handshake is not confirmed): if (handshake is not confirmed):
return pto_timeout, pto_space return pto_timeout, pto_space
// Include max_ack_delay and backoff for Application Data. // Include max_ack_delay and backoff for Application Data.
duration += max_ack_delay * (2 ^ pto_count) duration += max_ack_delay * (2 ^ pto_count)
t = time_of_last_ack_eliciting_packet[space] + duration t = time_of_last_ack_eliciting_packet[space] + duration
if (t < pto_timeout): if (t < pto_timeout):
skipping to change at page 39, line 22 skipping to change at line 1760
OnLossDetectionTimeout(): OnLossDetectionTimeout():
earliest_loss_time, pn_space = GetLossTimeAndSpace() earliest_loss_time, pn_space = GetLossTimeAndSpace()
if (earliest_loss_time != 0): if (earliest_loss_time != 0):
// Time threshold loss Detection // Time threshold loss Detection
lost_packets = DetectAndRemoveLostPackets(pn_space) lost_packets = DetectAndRemoveLostPackets(pn_space)
assert(!lost_packets.empty()) assert(!lost_packets.empty())
OnPacketsLost(lost_packets) OnPacketsLost(lost_packets)
SetLossDetectionTimer() SetLossDetectionTimer()
return return
if (bytes_in_flight > 0): if (no ack-eliciting packets in flight):
// PTO. Send new data if available, else retransmit old data.
// If neither is available, send a single PING frame.
_, pn_space = GetPtoTimeAndSpace()
SendOneOrTwoAckElicitingPackets(pn_space)
else:
assert(!PeerCompletedAddressValidation()) assert(!PeerCompletedAddressValidation())
// Client sends an anti-deadlock packet: Initial is padded // Client sends an anti-deadlock packet: Initial is padded
// to earn more anti-amplification credit, // to earn more anti-amplification credit,
// a Handshake packet proves address ownership. // a Handshake packet proves address ownership.
if (has Handshake keys): if (has Handshake keys):
SendOneAckElicitingHandshakePacket() SendOneAckElicitingHandshakePacket()
else: else:
SendOneAckElicitingPaddedInitialPacket() SendOneAckElicitingPaddedInitialPacket()
else:
// PTO. Send new data if available, else retransmit old data.
// If neither is available, send a single PING frame.
_, pn_space = GetPtoTimeAndSpace()
SendOneOrTwoAckElicitingPackets(pn_space)
pto_count++ pto_count++
SetLossDetectionTimer() SetLossDetectionTimer()
A.10. Detecting Lost Packets A.10. Detecting Lost Packets
DetectAndRemoveLostPackets is called every time an ACK is received or DetectAndRemoveLostPackets is called every time an ACK is received or
the time threshold loss detection timer expires. This function the time threshold loss detection timer expires. This function
operates on the sent_packets for that packet number space and returns operates on the sent_packets for that packet number space and returns
a list of packets newly detected as lost. a list of packets newly detected as lost.
skipping to change at page 41, line 13 skipping to change at line 1844
SetLossDetectionTimer() SetLossDetectionTimer()
Appendix B. Congestion Control Pseudocode Appendix B. Congestion Control Pseudocode
We now describe an example implementation of the congestion We now describe an example implementation of the congestion
controller described in Section 7. controller described in Section 7.
The pseudocode segments in this section are licensed as Code The pseudocode segments in this section are licensed as Code
Components; see the copyright notice. Components; see the copyright notice.
B.1. Constants of interest B.1. Constants of Interest
Constants used in congestion control are based on a combination of Constants used in congestion control are based on a combination of
RFCs, papers, and common practice. RFCs, papers, and common practice.
kInitialWindow: Default limit on the initial bytes in flight as kInitialWindow: Default limit on the initial bytes in flight as
described in Section 7.2. described in Section 7.2.
kMinimumWindow: Minimum congestion window in bytes as described in kMinimumWindow: Minimum congestion window in bytes as described in
Section 7.2. Section 7.2.
kLossReductionFactor: Scaling factor applied to reduce the kLossReductionFactor: Scaling factor applied to reduce the
congestion window when a new loss event is detected. Section 7 congestion window when a new loss event is detected. Section 7
recommends a value is 0.5. recommends a value of 0.5.
kPersistentCongestionThreshold: Period of time for persistent kPersistentCongestionThreshold: Period of time for persistent
congestion to be established, specified as a PTO multiplier. congestion to be established, specified as a PTO multiplier.
Section 7.6 recommends a value of 3. Section 7.6 recommends a value of 3.
B.2. Variables of interest B.2. Variables of Interest
Variables required to implement the congestion control mechanisms are Variables required to implement the congestion control mechanisms are
described in this section. described in this section.
max_datagram_size: The sender's current maximum payload size. Does max_datagram_size: The sender's current maximum payload size. This
not include UDP or IP overhead. The max datagram size is used for does not include UDP or IP overhead. The max datagram size is
congestion window computations. An endpoint sets the value of used for congestion window computations. An endpoint sets the
this variable based on its Path Maximum Transmission Unit (PMTU; value of this variable based on its Path Maximum Transmission Unit
see Section 14.2 of [QUIC-TRANSPORT]), with a minimum value of (PMTU; see Section 14.2 of [QUIC-TRANSPORT]), with a minimum value
1200 bytes. of 1200 bytes.
ecn_ce_counters[kPacketNumberSpace]: The highest value reported for ecn_ce_counters[kPacketNumberSpace]: The highest value reported for
the ECN-CE counter in the packet number space by the peer in an the ECN-CE counter in the packet number space by the peer in an
ACK frame. This value is used to detect increases in the reported ACK frame. This value is used to detect increases in the reported
ECN-CE counter. ECN-CE counter.
bytes_in_flight: The sum of the size in bytes of all sent packets bytes_in_flight: The sum of the size in bytes of all sent packets
that contain at least one ack-eliciting or PADDING frame, and have that contain at least one ack-eliciting or PADDING frame and have
not been acknowledged or declared lost. The size does not include not been acknowledged or declared lost. The size does not include
IP or UDP overhead, but does include the QUIC header and AEAD IP or UDP overhead, but does include the QUIC header and
overhead. Packets only containing ACK frames do not count towards Authenticated Encryption with Associated Data (AEAD) overhead.
Packets only containing ACK frames do not count toward
bytes_in_flight to ensure congestion control does not impede bytes_in_flight to ensure congestion control does not impede
congestion feedback. congestion feedback.
congestion_window: Maximum number of bytes allowed to be in flight. congestion_window: Maximum number of bytes allowed to be in flight.
congestion_recovery_start_time: The time the current recovery period congestion_recovery_start_time: The time the current recovery period
started due to the detection of loss or ECN. When a packet sent started due to the detection of loss or ECN. When a packet sent
after this time is acknowledged, QUIC exits congestion recovery. after this time is acknowledged, QUIC exits congestion recovery.
ssthresh: Slow start threshold in bytes. When the congestion window ssthresh: Slow start threshold in bytes. When the congestion window
skipping to change at page 42, line 35 skipping to change at line 1916
congestion_window = kInitialWindow congestion_window = kInitialWindow
bytes_in_flight = 0 bytes_in_flight = 0
congestion_recovery_start_time = 0 congestion_recovery_start_time = 0
ssthresh = infinite ssthresh = infinite
for pn_space in [ Initial, Handshake, ApplicationData ]: for pn_space in [ Initial, Handshake, ApplicationData ]:
ecn_ce_counters[pn_space] = 0 ecn_ce_counters[pn_space] = 0
B.4. On Packet Sent B.4. On Packet Sent
Whenever a packet is sent, and it contains non-ACK frames, the packet Whenever a packet is sent and it contains non-ACK frames, the packet
increases bytes_in_flight. increases bytes_in_flight.
OnPacketSentCC(sent_bytes): OnPacketSentCC(sent_bytes):
bytes_in_flight += sent_bytes bytes_in_flight += sent_bytes
B.5. On Packet Acknowledgment B.5. On Packet Acknowledgment
Invoked from loss detection's OnAckReceived and is supplied with the This is invoked from loss detection's OnAckReceived and is supplied
newly acked_packets from sent_packets. with the newly acked_packets from sent_packets.
In congestion avoidance, implementers that use an integer In congestion avoidance, implementers that use an integer
representation for congestion_window should be careful with division, representation for congestion_window should be careful with division
and can use the alternative approach suggested in Section 2.1 of and can use the alternative approach suggested in Section 2.1 of
[RFC3465]. [RFC3465].
InCongestionRecovery(sent_time): InCongestionRecovery(sent_time):
return sent_time <= congestion_recovery_start_time return sent_time <= congestion_recovery_start_time
OnPacketsAcked(acked_packets): OnPacketsAcked(acked_packets):
for acked_packet in acked_packets: for acked_packet in acked_packets:
OnPacketAcked(acked_packet) OnPacketAcked(acked_packet)
skipping to change at page 43, line 35 skipping to change at line 1962
// Slow start. // Slow start.
congestion_window += acked_packet.sent_bytes congestion_window += acked_packet.sent_bytes
else: else:
// Congestion avoidance. // Congestion avoidance.
congestion_window += congestion_window +=
max_datagram_size * acked_packet.sent_bytes max_datagram_size * acked_packet.sent_bytes
/ congestion_window / congestion_window
B.6. On New Congestion Event B.6. On New Congestion Event
Invoked from ProcessECN and OnPacketsLost when a new congestion event This is invoked from ProcessECN and OnPacketsLost when a new
is detected. If not already in recovery, this starts a recovery congestion event is detected. If not already in recovery, this
period and reduces the slow start threshold and congestion window starts a recovery period and reduces the slow start threshold and
immediately. congestion window immediately.
OnCongestionEvent(sent_time): OnCongestionEvent(sent_time):
// No reaction if already in a recovery period. // No reaction if already in a recovery period.
if (InCongestionRecovery(sent_time)): if (InCongestionRecovery(sent_time)):
return return
// Enter recovery period. // Enter recovery period.
congestion_recovery_start_time = now() congestion_recovery_start_time = now()
ssthresh = congestion_window * kLossReductionFactor ssthresh = congestion_window * kLossReductionFactor
congestion_window = max(ssthresh, kMinimumWindow) congestion_window = max(ssthresh, kMinimumWindow)
// A packet can be sent to speed up loss recovery. // A packet can be sent to speed up loss recovery.
MaybeSendOnePacket() MaybeSendOnePacket()
B.7. Process ECN Information B.7. Process ECN Information
Invoked when an ACK frame with an ECN section is received from the This is invoked when an ACK frame with an ECN section is received
peer. from the peer.
ProcessECN(ack, pn_space): ProcessECN(ack, pn_space):
// If the ECN-CE counter reported by the peer has increased, // If the ECN-CE counter reported by the peer has increased,
// this could be a new congestion event. // this could be a new congestion event.
if (ack.ce_counter > ecn_ce_counters[pn_space]): if (ack.ce_counter > ecn_ce_counters[pn_space]):
ecn_ce_counters[pn_space] = ack.ce_counter ecn_ce_counters[pn_space] = ack.ce_counter
sent_time = sent_packets[ack.largest_acked].time_sent sent_time = sent_packets[ack.largest_acked].time_sent
OnCongestionEvent(sent_time) OnCongestionEvent(sent_time)
B.8. On Packets Lost B.8. On Packets Lost
Invoked when DetectAndRemoveLostPackets deems packets lost. This is invoked when DetectAndRemoveLostPackets deems packets lost.
OnPacketsLost(lost_packets): OnPacketsLost(lost_packets):
sent_time_of_last_loss = 0 sent_time_of_last_loss = 0
// Remove lost packets from bytes_in_flight. // Remove lost packets from bytes_in_flight.
for lost_packet in lost_packets: for lost_packet in lost_packets:
if lost_packet.in_flight: if lost_packet.in_flight:
bytes_in_flight -= lost_packet.sent_bytes bytes_in_flight -= lost_packet.sent_bytes
sent_time_of_last_loss = sent_time_of_last_loss =
max(sent_time_of_last_loss, lost_packet.time_sent) max(sent_time_of_last_loss, lost_packet.time_sent)
// Congestion event if in-flight packets were lost // Congestion event if in-flight packets were lost
skipping to change at page 44, line 47 skipping to change at line 2021
if (first_rtt_sample == 0): if (first_rtt_sample == 0):
return return
pc_lost = [] pc_lost = []
for lost in lost_packets: for lost in lost_packets:
if lost.time_sent > first_rtt_sample: if lost.time_sent > first_rtt_sample:
pc_lost.insert(lost) pc_lost.insert(lost)
if (InPersistentCongestion(pc_lost)): if (InPersistentCongestion(pc_lost)):
congestion_window = kMinimumWindow congestion_window = kMinimumWindow
congestion_recovery_start_time = 0 congestion_recovery_start_time = 0
B.9. Removing Discarded Packets From Bytes In Flight B.9. Removing Discarded Packets from Bytes in Flight
When Initial or Handshake keys are discarded, packets sent in that When Initial or Handshake keys are discarded, packets sent in that
space no longer count toward bytes in flight. space no longer count toward bytes in flight.
Pseudocode for RemoveFromBytesInFlight follows: Pseudocode for RemoveFromBytesInFlight follows:
RemoveFromBytesInFlight(discarded_packets): RemoveFromBytesInFlight(discarded_packets):
// Remove any unacknowledged packets from flight. // Remove any unacknowledged packets from flight.
foreach packet in discarded_packets: foreach packet in discarded_packets:
if packet.in_flight if packet.in_flight
bytes_in_flight -= size bytes_in_flight -= size
Appendix C. Change Log Contributors
*RFC Editor's Note:* Please remove this section prior to
publication of a final version of this document.
Issue and pull request numbers are listed with a leading octothorp.
C.1. Since draft-ietf-quic-recovery-32
* Clarifications to definition of persistent congestion (#4413,
#4414, #4421, #4429, #4437)
C.2. Since draft-ietf-quic-recovery-31
* Limit the number of Initial packets sent in response to
unauthenticated packets (#4183, #4188)
C.3. Since draft-ietf-quic-recovery-30
Editorial changes only.
C.4. Since draft-ietf-quic-recovery-29
* Allow caching of packets that can't be decrypted, by allowing the
reported acknowledgment delay to exceed max_ack_delay prior to
confirming the handshake (#3821, #3980, #4035, #3874)
* Persistent congestion cannot include packets sent before the first
RTT sample for the path (#3875, #3889)
* Recommend reset of min_rtt in persistent congestion (#3927, #3975)
* Persistent congestion is independent of packet number space
(#3939, #3961)
* Only limit bursts to the initial window without information about
the path (#3892, #3936)
* Add normative requirements for increasing and reducing the
congestion window (#3944, #3978, #3997, #3998)
C.5. Since draft-ietf-quic-recovery-28
* Refactored pseudocode to correct PTO calculation (#3564, #3674,
#3681)
C.6. Since draft-ietf-quic-recovery-27
* Added recommendations for speeding up handshake under some loss
conditions (#3078, #3080)
* PTO count is reset when handshake progress is made (#3272, #3415)
* PTO count is not reset by a client when the server might be
awaiting address validation (#3546, #3551)
* Recommend repairing losses immediately after entering the recovery
period (#3335, #3443)
* Clarified what loss conditions can be ignored during the handshake
(#3456, #3450)
* Allow, but don't recommend, using RTT from previous connection to
seed RTT (#3464, #3496)
* Recommend use of adaptive loss detection thresholds (#3571, #3572)
C.7. Since draft-ietf-quic-recovery-26
No changes.
C.8. Since draft-ietf-quic-recovery-25
No significant changes.
C.9. Since draft-ietf-quic-recovery-24
* Require congestion control of some sort (#3247, #3244, #3248)
* Set a minimum reordering threshold (#3256, #3240)
* PTO is specific to a packet number space (#3067, #3074, #3066)
C.10. Since draft-ietf-quic-recovery-23
* Define under-utilizing the congestion window (#2630, #2686, #2675)
* PTO MUST send data if possible (#3056, #3057)
* Connection Close is not ack-eliciting (#3097, #3098)
* MUST limit bursts to the initial congestion window (#3160)
* Define the current max_datagram_size for congestion control
(#3041, #3167)
C.11. Since draft-ietf-quic-recovery-22
* PTO should always send an ack-eliciting packet (#2895)
* Unify the Handshake Timer with the PTO timer (#2648, #2658, #2886)
* Move ACK generation text to transport draft (#1860, #2916)
C.12. Since draft-ietf-quic-recovery-21
* No changes
C.13. Since draft-ietf-quic-recovery-20
* Path validation can be used as initial RTT value (#2644, #2687)
* max_ack_delay transport parameter defaults to 0 (#2638, #2646)
* ACK delay only measures intentional delays induced by the
implementation (#2596, #2786)
C.14. Since draft-ietf-quic-recovery-19
* Change kPersistentThreshold from an exponent to a multiplier
(#2557)
* Send a PING if the PTO timer fires and there's nothing to send
(#2624)
* Set loss delay to at least kGranularity (#2617)
* Merge application limited and sending after idle sections. Always
limit burst size instead of requiring resetting CWND to initial
CWND after idle (#2605)
* Rewrite RTT estimation, allow RTT samples where a newly acked
packet is ack-eliciting but the largest_acked is not (#2592)
* Don't arm the handshake timer if there is no handshake data
(#2590)
* Clarify that the time threshold loss alarm takes precedence over
the crypto handshake timer (#2590, #2620)
* Change initial RTT to 500ms to align with RFC6298 (#2184)
C.15. Since draft-ietf-quic-recovery-18
* Change IW byte limit to 14720 from 14600 (#2494)
* Update PTO calculation to match RFC6298 (#2480, #2489, #2490)
* Improve loss detection's description of multiple packet number
spaces and pseudocode (#2485, #2451, #2417)
* Declare persistent congestion even if non-probe packets are sent
and don't make persistent congestion more aggressive than RTO
verified was (#2365, #2244)
* Move pseudocode to the appendices (#2408)
* What to send on multiple PTOs (#2380)
C.16. Since draft-ietf-quic-recovery-17
* After Probe Timeout discard in-flight packets or send another
(#2212, #1965)
* Endpoints discard initial keys as soon as handshake keys are
available (#1951, #2045)
* 0-RTT state is discarded when 0-RTT is rejected (#2300)
* Loss detection timer is cancelled when ack-eliciting frames are in
flight (#2117, #2093)
* Packets are declared lost if they are in flight (#2104)
* After becoming idle, either pace packets or reset the congestion
controller (#2138, 2187)
* Process ECN counts before marking packets lost (#2142)
* Mark packets lost before resetting crypto_count and pto_count
(#2208, #2209)
* Congestion and loss recovery state are discarded when keys are
discarded (#2327)
C.17. Since draft-ietf-quic-recovery-16
* Unify TLP and RTO into a single PTO; eliminate min RTO, min TLP
and min crypto timeouts; eliminate timeout validation (#2114,
#2166, #2168, #1017)
* Redefine how congestion avoidance in terms of when the period
starts (#1928, #1930)
* Document what needs to be tracked for packets that are in flight
(#765, #1724, #1939)
* Integrate both time and packet thresholds into loss detection
(#1969, #1212, #934, #1974)
* Reduce congestion window after idle, unless pacing is used (#2007,
#2023)
* Disable RTT calculation for packets that don't elicit
acknowledgment (#2060, #2078)
* Limit ack_delay by max_ack_delay (#2060, #2099)
* Initial keys are discarded once Handshake keys are available
(#1951, #2045)
* Reorder ECN and loss detection in pseudocode (#2142)
* Only cancel loss detection timer if ack-eliciting packets are in
flight (#2093, #2117)
C.18. Since draft-ietf-quic-recovery-14
* Used max_ack_delay from transport params (#1796, #1782)
* Merge ACK and ACK_ECN (#1783)
C.19. Since draft-ietf-quic-recovery-13
* Corrected the lack of ssthresh reduction in CongestionEvent
pseudocode (#1598)
* Considerations for ECN spoofing (#1426, #1626)
* Clarifications for PADDING and congestion control (#837, #838,
#1517, #1531, #1540)
* Reduce early retransmission timer to RTT/8 (#945, #1581)
* Packets are declared lost after an RTO is verified (#935, #1582)
C.20. Since draft-ietf-quic-recovery-12
* Changes to manage separate packet number spaces and encryption
levels (#1190, #1242, #1413, #1450)
* Added ECN feedback mechanisms and handling; new ACK_ECN frame
(#804, #805, #1372)
C.21. Since draft-ietf-quic-recovery-11
No significant changes.
C.22. Since draft-ietf-quic-recovery-10
* Improved text on ack generation (#1139, #1159)
* Make references to TCP recovery mechanisms informational (#1195)
* Define time_of_last_sent_handshake_packet (#1171)
* Added signal from TLS the data it includes needs to be sent in a
Retry packet (#1061, #1199)
* Minimum RTT (min_rtt) is initialized with an infinite value
(#1169)
C.23. Since draft-ietf-quic-recovery-09
No significant changes.
C.24. Since draft-ietf-quic-recovery-08
* Clarified pacing and RTO (#967, #977)
C.25. Since draft-ietf-quic-recovery-07
* Include ACK delay in RTO(and TLP) computations (#981)
* ACK delay in SRTT computation (#961)
* Default RTT and Slow Start (#590)
* Many editorial fixes.
C.26. Since draft-ietf-quic-recovery-06
No significant changes.
C.27. Since draft-ietf-quic-recovery-05
* Add more congestion control text (#776)
C.28. Since draft-ietf-quic-recovery-04
No significant changes.
C.29. Since draft-ietf-quic-recovery-03
No significant changes.
C.30. Since draft-ietf-quic-recovery-02
* Integrate F-RTO (#544, #409)
* Add congestion control (#545, #395)
* Require connection abort if a skipped packet was acknowledged
(#415)
* Simplify RTO calculations (#142, #417)
C.31. Since draft-ietf-quic-recovery-01
* Overview added to loss detection
* Changes initial default RTT to 100ms
* Added time-based loss detection and fixes early retransmit
* Clarified loss recovery for handshake packets
* Fixed references and made TCP references informative
C.32. Since draft-ietf-quic-recovery-00
* Improved description of constants and ACK behavior
C.33. Since draft-iyengar-quic-loss-recovery-01
* Adopted as base for draft-ietf-quic-recovery
* Updated authors/editors list
* Added table of contents
Appendix D. Contributors
The IETF QUIC Working Group received an enormous amount of support The IETF QUIC Working Group received an enormous amount of support
from many people. The following people provided substantive from many people. The following people provided substantive
contributions to this document: contributions to this document:
* Alessandro Ghedini * Alessandro Ghedini
* Benjamin Saunders * Benjamin Saunders
* Gorry Fairhurst * Gorry Fairhurst
* 山本和彦 (Kazu Yamamoto) * 山本和彦 (Kazu Yamamoto)
* 奥 一穂 (Kazuho Oku) * 奥 一穂 (Kazuho Oku)
* Lars Eggert * Lars Eggert
* Magnus Westerlund * Magnus Westerlund
* Marten Seemann * Marten Seemann
* Martin Duke * Martin Duke
* Martin Thomson * Martin Thomson
* Mirja Kühlewind * Mirja Kühlewind
* Nick Banks * Nick Banks
* Praveen Balasubramanian * Praveen Balasubramanian
Acknowledgments
Authors' Addresses Authors' Addresses
Jana Iyengar (editor) Jana Iyengar (editor)
Fastly Fastly
Email: jri.ietf@gmail.com Email: jri.ietf@gmail.com
Ian Swett (editor) Ian Swett (editor)
Google Google
 End of changes. 166 change blocks. 
790 lines changed or deleted 416 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/