rfc9840.original   rfc9840.txt 
Network Working Group M. Bagnulo Internet Research Task Force (IRTF) M. Bagnulo
Internet-Draft A. Garcia-Martinez Request for Comments: 9840 A. García-Martínez
Intended status: Experimental Universidad Carlos III de Madrid Category: Experimental Universidad Carlos III de Madrid
Expires: 6 August 2025 G. Montenegro ISSN: 2070-1721 G. Montenegro
P. Balasubramanian P. Balasubramanian
Confluent Confluent
2 February 2025 September 2025
rLEDBAT: receiver-driven Low Extra Delay Background Transport for TCP rLEDBAT: Receiver-Driven Low Extra Delay Background Transport for TCP
draft-irtf-iccrg-rledbat-10
Abstract Abstract
This document specifies rLEDBAT, a set of mechanisms that enable the This document specifies receiver-driven Low Extra Delay Background
execution of a less-than-best-effort congestion control algorithm for Transport (rLEDBAT) -- a set of mechanisms that enable the execution
TCP at the receiver end. This document is a product of the Internet of a less-than-best-effort congestion control algorithm for TCP at
the receiver end. This document is a product of the Internet
Congestion Control Research Group (ICCRG) of the Internet Research Congestion Control Research Group (ICCRG) of the Internet Research
Task Force (IRTF). Task Force (IRTF).
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the This document is not an Internet Standards Track specification; it is
provisions of BCP 78 and BCP 79. published for examination, experimental implementation, and
evaluation.
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 defines an Experimental Protocol for the Internet
and may be updated, replaced, or obsoleted by other documents at any community. This document is a product of the Internet Research Task
time. It is inappropriate to use Internet-Drafts as reference Force (IRTF). The IRTF publishes the results of Internet-related
material or to cite them other than as "work in progress." research and development activities. These results might not be
suitable for deployment. This RFC represents the consensus of the
Internet Congestion Control Research Group of the Internet Research
Task Force (IRTF). Documents approved for publication by the IRSG
are not candidates for any level of Internet Standard; see Section 2
of RFC 7841.
This Internet-Draft will expire on 6 August 2025. 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/rfc9840.
Copyright Notice Copyright Notice
Copyright (c) 2025 IETF Trust and the persons identified as the Copyright (c) 2025 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 Revised BSD License text as to this document.
described in Section 4.e of the Trust Legal Provisions and are
provided without warranty as described in the Revised BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1. Introduction
2. Motivations for rLEDBAT . . . . . . . . . . . . . . . . . . . 3 2. Conventions and Terminology
3. rLEDBAT mechanisms . . . . . . . . . . . . . . . . . . . . . 4 3. Motivations for rLEDBAT
3.1. Controlling the receive window . . . . . . . . . . . . . 6 4. rLEDBAT Mechanisms
3.1.1. Avoiding window shrinking . . . . . . . . . . . . . . 7 4.1. Controlling the Receive Window
3.1.2. Setting the Window Scale Option . . . . . . . . . . . 8 4.1.1. Avoiding Window Shrinking
3.2. Measuring delays . . . . . . . . . . . . . . . . . . . . 8 4.1.2. Setting the Window Scale Option
3.2.1. Measuring RTT to estimate the queueing delay . . . . 9 4.2. Measuring Delays
3.2.2. Measuring one way delay to estimate the queueing 4.2.1. Measuring RTT to Estimate the Queuing Delay
delay . . . . . . . . . . . . . . . . . . . . . . . . 11 4.2.2. Measuring One-Way Delay to Estimate the Queuing Delay
3.3. Detecting packet losses and retransmissions . . . . . . . 13 4.3. Detecting Packet Losses and Retransmissions
4. Experiment Considerations . . . . . . . . . . . . . . . . . . 13 5. Experiment Considerations
4.1. Status of the experiment at the time of this writing. . . 14 5.1. Status of the Experiment at the Time of This Writing
5. Security Considerations . . . . . . . . . . . . . . . . . . . 15 6. Security Considerations
6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 16 7. IANA Considerations
7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 16 8. References
8. Informative References . . . . . . . . . . . . . . . . . . . 16 8.1. Normative References
Appendix A. Terminology . . . . . . . . . . . . . . . . . . . . 17 8.2. Informative References
Appendix B. rLEDBAT pseudo-code . . . . . . . . . . . . . . . . 18 Appendix A. rLEDBAT Pseudocode
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 20 Acknowledgments
Authors' Addresses
1. Introduction 1. Introduction
LEDBAT (Low Extra Delay Background Transport) [RFC6817] is a LEDBAT (Low Extra Delay Background Transport) [RFC6817] is a
congestion-control algorithm used for less-than-best-effort (LBE) congestion control algorithm used for less-than-best-effort (LBE)
traffic. traffic.
When LEDBAT traffic shares a bottleneck with other traffic using When LEDBAT traffic shares a bottleneck with other traffic using
standard congestion control algorithms (for example, TCP traffic standard congestion control algorithms (for example, TCP traffic
using Cubic[RFC9438], hereafter referred as standard-TCP for short), using CUBIC [RFC9438], hereafter referred to as "standard-TCP" for
it reduces its sending rate earlier and more aggressively than short), it reduces its sending rate earlier and more aggressively
standard-TCP congestion control, allowing other non-background than standard-TCP congestion control, allowing other non-background
traffic to use more of the available capacity. In the absence of traffic to use more of the available capacity. In the absence of
competing traffic, LEDBAT aims to make an efficient use of the competing traffic, LEDBAT aims to make efficient use of the available
available capacity, while keeping the queuing delay within predefined capacity, while keeping the queuing delay within predefined bounds.
bounds.
LEDBAT reacts both to packet loss and to variations in delay. With LEDBAT reacts to both packet loss and variations in delay. With
respect to packet loss, LEDBAT reacts with a multiplicative decrease, respect to packet loss, LEDBAT reacts with a multiplicative decrease,
similar to most TCP congestion controllers. Regarding delay, LEDBAT similar to most TCP congestion controllers. Regarding delay, LEDBAT
aims for a target queueing delay. When the measured current queueing aims for a target queuing delay. When the measured current queuing
delay is below the target, LEDBAT increases the sending rate and when delay is below the target, LEDBAT increases the sending rate, and
the delay is above the target, it reduces the sending rate. LEDBAT when the delay is above the target, it reduces the sending rate.
estimates the queuing delay by subtracting the measured current one- LEDBAT estimates the queuing delay by subtracting the measured
way delay from the estimated base one-way delay (i.e. the one-way current one-way delay from the estimated base one-way delay (i.e.,
delay in the absence of queues). the one-way delay in the absence of queues).
The LEDBAT specification [RFC6817] defines the LEDBAT congestion- The LEDBAT specification [RFC6817] defines the LEDBAT congestion
control algorithm, implemented in the sender to control its sending control algorithm, implemented in the sender to control its sending
rate. LEDBAT is specified in a protocol and layer agnostic manner. rate. LEDBAT is specified in a protocol-agnostic and layer-agnostic
manner.
LEDBAT++ [I-D.irtf-iccrg-ledbat-plus-plus] is also an LBE congestion LEDBAT++ [LEDBAT++] is also an LBE congestion control algorithm that
control algorithm which is inspired by LEDBAT while addressing is inspired by LEDBAT while addressing several problems identified
several problems identified with the original LEDBAT specification. with the original LEDBAT specification. In particular, the
In particular the differences between LEDBAT and LEDBAT++ include: i) differences between LEDBAT and LEDBAT++ include the following:
LEDBAT++ uses the round-trip-time (RTT) (as opposed to the one way
delay used in LEDBAT) to estimate the queuing delay; ii) LEDBAT++
uses an Additive Increase/Multiplicative Decrease algorithm to
achieve inter-LEDBAT++ fairness and avoid the late-comer advantage
observed in LEDBAT; iii) LEDBAT++ performs periodic slowdowns to
improve the measurement of the base delay; iv) LEDBAT++ is defined
for TCP.
In this specification, we describe rLEDBAT, a set of mechanisms that i) LEDBAT++ uses the round-trip time (RTT) (as opposed to the one-
enable the execution of an LBE delay-based congestion control way delay used in LEDBAT) to estimate the queuing delay.
algorithm such as LEDBAT or LEDBAT++ at the receiver end of a TCP
connection. ii) LEDBAT++ uses an additive increase/multiplicative decrease
algorithm to achieve inter-LEDBAT++ fairness and avoid the
latecomer advantage observed in LEDBAT.
iii) LEDBAT++ performs periodic slowdowns to improve the measurement
of the base delay.
iv) LEDBAT++ is defined for TCP.
In this specification, we describe receiver-driven Low Extra Delay
Background Transport (rLEDBAT) -- a set of mechanisms that enable the
execution of an LBE delay-based congestion control algorithm such as
LEDBAT or LEDBAT++ at the receiver end of a TCP connection.
The consensus of the Internet Congestion Control Research Group The consensus of the Internet Congestion Control Research Group
(ICCRG) is to publish this document to encourage further (ICCRG) is to publish this document to encourage further
experimentation and review of rLEDBAT. This document is not an IETF experimentation and review of rLEDBAT. This document is not an IETF
product and is not a standard. The status of this document is product and is not an Internet Standards Track specification. The
experimental. In section 4 titled Experiment Considerations, we status of this document is Experimental. In Section 5 ("Experiment
describe the purpose of the experiment and its current status. Considerations"), we describe the purpose of the experiment and its
current status.
2. Motivations for rLEDBAT 2. Conventions and Terminology
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in
BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all
capitals, as shown here.
We use the following abbreviations throughout the text and include
them here for the reader's convenience:
RCV.WND: The value included in the Receive Window field of the TCP
header (the computation of which is modified by its
specification).
SND.WND: The TCP sender's window.
cwnd: The congestion window as computed by the congestion control
algorithm running at the TCP sender.
RLWND: The window value calculated by the rLEDBAT algorithm.
fcwnd: The value that a standard-TCP receiver compliant with
[RFC9293] calculates to set in the receive window for flow control
purposes.
RCV.HGH: The highest sequence number corresponding to a received
byte of data at one point in time.
TSV.HGH: The Timestamp Value (TSval) [RFC7323] corresponding to the
segment in which RCV.HGH was carried at that point in time.
SEG.SEQ: The sequence number of the last received segment.
TSV.SEQ: The TSval of the last received segment.
3. Motivations for rLEDBAT
rLEDBAT enables new use cases and new deployment models, fostering rLEDBAT enables new use cases and new deployment models, fostering
the use of LBE traffic. The following scenarios are enabled by the use of LBE traffic. The following scenarios are enabled by
rLEDBAT: rLEDBAT:
Content Delivery Networks and more sophisticated file distribution Content Delivery Networks (CDNs) and more sophisticated file
scenarios: Consider the case where the source of a file to be distribution scenarios:
distributed (e.g., a software developer that wishes to distribute Consider the case where the source of a file to be distributed
a software update) would prefer to use LBE and it enables LEDBAT/ (e.g., a software developer that wishes to distribute a software
LEDBAT++ in the servers containing the source file. However, update) would prefer to use LBE and enables LEDBAT/LEDBAT++ in the
because the file is being distributed through a CDN that does not servers containing the source file. However, because the file is
implement LBE congestion control, the result is that the file being distributed through a CDN that does not implement LBE
transfers originated from CDN surrogates will not be using LBE. congestion control, the result is that the file transfers
originated from CDN surrogates will not be using LBE.
Interestingly enough, in the case of the software update, the Interestingly enough, in the case of the software update, the
developer may also control the software performing the download in developer may also control the software performing the download in
the client, the receiver of the file, but because current LEDBAT/ the client (the receiver of the file), but because current LEDBAT/
LEDBAT++ are sender-based algorithms, controlling the client is LEDBAT++ are sender-based algorithms, controlling the client is
not enough to enable LBE congestion control in the communication. not enough to enable LBE congestion control in the
rLEDBAT would enable the use of LBE traffic class for file communication. rLEDBAT would enable the use of an LBE traffic
distribution in this setup. class for file distribution in this setup.
Interference from proxies and other middleboxes: Proxies and other Interference from proxies and other middleboxes:
middleboxes are commonplace in the Internet. For instance, in the Proxies and other middleboxes are commonplace in the Internet.
case of mobile networks, proxies are frequently used. In the case For instance, in the case of mobile networks, proxies are
of enterprise networks, it is common to deploy corporate proxies frequently used. In the case of enterprise networks, it is common
for filtering and firewalling. In the case of satellite links, to deploy corporate proxies for filtering and firewalling. In the
Performance Enhancement Proxies (PEPs) are deployed to mitigate case of satellite links, Performance Enhancing Proxies (PEPs) are
the effect of the long delay in TCP connection. These proxies deployed to mitigate the effect of long delays in a TCP
terminate the TCP connection on both ends and prevent the use of connection. These proxies terminate the TCP connection on both
LBE congestion control in the segment between the proxy and the ends and prevent the use of LBE congestion control in the segment
sink of the content, the client. By enabling rLEDBAT, clients between the proxy and the sink of the content, the client. By
would be able to enable LBE traffic between them and the proxy. enabling rLEDBAT, clients can then enable LBE traffic between them
and the proxy.
Receiver-defined preferences. It is frequent that the bottleneck Receiver-defined preferences:
of the communication is the access link. This is particularly Frequently, the access link is the communication bottleneck. This
true in the case of mobile devices. It is then especially is particularly true in the case of mobile devices. It is then
relevant for mobile devices to properly manage the capacity of the especially relevant for mobile devices to properly manage the
access link. With current technologies, it is possible for the capacity of the access link. With current technologies, it is
mobile device to use different congestion control algorithms possible for the mobile device to use different congestion control
expressing different preferences for the traffic. For instance, a algorithms expressing different preferences for the traffic. For
device can choose to use standard-TCP for some traffic and to use instance, a device can choose to use standard-TCP for some traffic
LEDBAT/LEDBAT++ for other traffic. However, this would only and use LEDBAT/LEDBAT++ for other traffic. However, this would
affect the outgoing traffic since both standard-TCP and LEDBAT/ only affect the outgoing traffic, since both standard-TCP and
LEDBAT++ are sender-driven. The mobile device has no means to LEDBAT/LEDBAT++ are driven by the sender. The mobile device has
manage the traffic in the down-link, which is in most cases, the no means to manage the traffic in the downlink, which is, in most
communication bottleneck for a typical eye-ball end-user. rLEDBAT cases, the communication bottleneck for a typical "eyeball" end
enables the mobile device to selectively use LBE traffic class for user. rLEDBAT enables the mobile device to selectively use an LBE
some of the incoming traffic. For instance, by using rLEDBAT, a traffic class for some of the incoming traffic. For instance, by
user can use regular standard-TCP/UDP for video stream (e.g., using rLEDBAT, a user can use regular standard-TCP/UDP for a video
Youtube) and use rLEDBAT for other background file download. stream (e.g., YouTube) and use rLEDBAT for other background file
downloads.
3. rLEDBAT mechanisms 4. rLEDBAT Mechanisms
rLEDBAT provides the mechanisms to implement an LBE congestion rLEDBAT provides the mechanisms to implement an LBE congestion
control algorithm at the receiver-end of a TCP connection. The control algorithm at the receiver end of a TCP connection. The
rLEDBAT receiver controls the sender's rate through the Receive rLEDBAT receiver controls the sender's rate through the receive
Window announced by the receiver in the TCP header. window announced by the receiver in the TCP header.
rLEDBAT assumes that the sender is a standard TCP sender. rLEDBAT rLEDBAT assumes that the sender is a standard-TCP sender. rLEDBAT
does not require any rLEDBAT-specific modifications to the TCP does not require any rLEDBAT-specific modifications to the TCP
sender. The envisioned deployment model for rLEDBAT is that the sender. The envisioned deployment model for rLEDBAT is that the
clients implement rLEDBAT and this enables rLEDBAT in communications clients implement rLEDBAT and this enables rLEDBAT in communications
with existent standard TCP senders. In particular, the sender MUST with existing standard-TCP senders. In particular, the sender MUST
implement [RFC9293] and it also MUST implement the Time Stamp Option implement [RFC9293] and also MUST implement the TCP Timestamps (TS)
as defined in [RFC7323]. Also, the sender should implement some of option as defined in [RFC7323]. Also, the sender should implement
the standard congestion control mechanisms, such as Cubic [RFC9438] some of the standard congestion control mechanisms, such as CUBIC
or New Reno [RFC5681]. [RFC9438] or NewReno [RFC5681] [RFC6582].
rLEDBAT does not define a new congestion control algorithm. The LBE rLEDBAT does not define a new congestion control algorithm. The
congestion control algorithm executed in the rLEDBAT receiver is definition of the actual LBE congestion control algorithm executed in
defined in other documents. The rLEDBAT receiver MUST use an LBE the rLEDBAT receiver is beyond the scope of this document. The
congestion control algorithm. Because rLEDBAT assumes a standard TCP rLEDBAT receiver MUST use an LBE congestion control algorithm.
sender, the sender will be using a "best effort" congestion control Because rLEDBAT assumes a standard-TCP sender, the sender will be
algorithm (such as Cubic or New Reno). Since rLEDBAT uses the using a "best effort" congestion control algorithm (such as CUBIC or
Receive Window to control the sender's rate and the sender calculates NewReno). Since rLEDBAT uses the receive window to control the
the sender's window as the minimum of the Receive window and the sender's rate and the sender calculates the sender's window as the
congestion window, rLEDBAT will only be effective as long as the minimum of the receive window and the congestion window, rLEDBAT will
congestion control algorithm executed in the receiver yields a only be effective as long as the congestion control algorithm
smaller window than the one calculated by the sender. This is executed in the receiver yields a smaller window than the one
normally the case when the receiver is using an LBE congestion calculated by the sender. This is normally the case when the
control algorithm. The rLEDBAT receiver SHOULD use the LEDBAT receiver is using an LBE congestion control algorithm. The rLEDBAT
congestion control algorithm [RFC6817] or the LEDBAT++ congestion receiver SHOULD use the LEDBAT congestion control algorithm [RFC6817]
control algorithm [I-D.irtf-iccrg-ledbat-plus-plus]. The rLEDBAT MAY or the LEDBAT++ congestion control algorithm [LEDBAT++]. rLEDBAT MAY
use other LBE congestion control algorithms defined elsewhere. use other LBE congestion control algorithms defined elsewhere.
Irrespective of which congestion control algorithm is executed in the Irrespective of which congestion control algorithm is executed in the
receiver, an rLEDBAT connection will never be more aggressive than receiver, a rLEDBAT connection will never be more aggressive than
standard-TCP since it is always bounded by the congestion control standard-TCP, since it is always bounded by the congestion control
algorithm executed at the sender. algorithm executed at the sender.
rLEDBAT is essentially composed of three types of mechanisms, namely, rLEDBAT is essentially composed of three types of mechanisms, namely
those that provide the means to measure the packet delay (either the those that provide the means to measure the packet delay (either the
round trip time or the one way delay, depending on the selected RTT or the one-way delay, depending on the selected algorithm),
algorithm), mechanisms to detect packet loss and the means to mechanisms to detect packet loss, and the means to manipulate the
manipulate the Receive Window to control the sender's rate. The receive window to control the sender's rate. The first two provide
former provide input to the LBE congestion control algorithm while input to the LBE congestion control algorithm, while the third uses
the latter uses the congestion window computed by the LBE congestion the congestion window computed by the LBE congestion control
control algorithm to manipulate the Receive window, as depicted in algorithm to manipulate the receive window, as depicted in Figure 1.
the figure.
+------------------------------------------+ +------------------------------------------+
| TCP receiver | | TCP Receiver |
| +-----------------+ | | +-----------------+ |
| | +------------+ | | | | +------------+ | |
| +---------------------| RTT | | | | +---------------------| RTT | | |
| | | | Estimation | | | | | | | Estimation | | |
| | | +------------+ | | | | | +------------+ | |
| | | | | | | | | |
| | | +------------+ | | | | | +------------+ | |
| | +--------------| Loss, RTX | | | | | +--------------| Loss, RTX | | |
| | | | | Detection | | | | | | | | Detection | | |
| | | | +------------+ | | | | | | +------------+ | |
| v v | | | | v v | | |
| +----------------+ | | | | +----------------+ | | |
| | LBE Congestion | | rLEDBAT | | | | LBE Congestion | | rLEDBAT | |
| | Control | | | | | | Control | | | |
| +----------------+ | | | | +----------------+ | | |
| | | +------------+ | | | | | +------------+ | |
| | | | RCV-WND | | | | | | | RCV.WND | | |
| +---------------->| Control | | | | +---------------->| Control | | |
| | +------------+ | | | | +------------+ | |
| +-----------------+ | | +-----------------+ |
+------------------------------------------+ +------------------------------------------+
Figure 1: The rLEDBAT architecture. Figure 1: The rLEDBAT Architecture
We describe each of the rLEDBAT components next. We next describe each of the rLEDBAT components.
3.1. Controlling the receive window 4.1. Controlling the Receive Window
rLEDBAT uses the Receive Window (RCV.WND) of TCP to enable the rLEDBAT uses the TCP receive window (RCV.WND) to enable the receiver
receiver to control the sender's rate. [RFC9293] defines that the to control the sender's rate. [RFC9293] specifies that the RCV.WND
RCV.WND is used to announce the available receive buffer to the is used to announce the available receive buffer to the sender for
sender for flow control purposes. In order to avoid confusion, we flow control purposes. In order to avoid confusion, we will call
will call fcwnd the value that a standard RFC793bis TCP receiver fcwnd the value that a standard-TCP receiver compliant with [RFC9293]
calculates to set in the receive window for flow control purposes. calculates to set in the receive window for flow control purposes.
We call RLWND the window value calculated by rLEDBAT algorithm and we We call RLWND the window value calculated by the rLEDBAT algorithm,
call RCV.WND the value actually included in the Receive Window field and we call RCV.WND the value actually included in the Receive Window
of the TCP header. For a RFC793bis receiver, RCV.WND == fcwnd. field of the TCP header. For a receiver compliant with [RFC9293],
RCV.WND == fcwnd.
In the case of rLEDBAT receiver, the rLEDBAT receiver MUST NOT set In the case of the rLEDBAT receiver, this receiver MUST NOT set the
the RCV.WND to a value larger than fcwnd and it SHOULD set the RCV.WND to a value larger than fcwnd and SHOULD set the RCV.WND to
RCV.WND to the minimum of RLWND and fcwnd, honoring both. the minimum of RLWND and fcwnd, honoring both.
When using rLEDBAT, two congestion controllers are in action in the When using rLEDBAT, two congestion controllers are in action in the
flow of data from the sender to the receiver, namely, the congestion flow of data from the sender to the receiver, namely the TCP
control algorithm of TCP in the sender side and the LBE congestion congestion control algorithm on the sender side and the LBE
control algorithm executed in the receiver and conveyed to the sender congestion control algorithm executed in the receiver and conveyed to
through the RCV.WND. In the normal TCP operation, the sender uses the sender through the RCV.WND. In the normal TCP operation, the
the minimum of the congestion window cwnd and the receiver window sender uses the minimum of the cwnd and the RCV.WND to calculate the
RCV.WND to calculate the sender's window SND.WND. This is also true SND.WND. This is also true for rLEDBAT, as the sender is a regular
for rLEDBAT, as the sender is a regular TCP sender. This guarantees TCP sender. This guarantees that the rLEDBAT flow will never
that the rLEDBAT flow will never transmit more aggressively than a transmit more aggressively than a standard-TCP flow, as the sender's
standard-TCP flow, as the sender's congestion window limits the congestion window limits the sending rate. Moreover, because an LBE
sending rate. Moreover, because a LBE congestion control algorithm congestion control algorithm such as LEDBAT/LEDBAT++ is designed to
such as LEDBAT/LEDBAT++ is designed to react earlier and more react earlier and more aggressively to congestion than regular TCP
aggressively to congestion than regular TCP congestion control, the congestion control, the RLWND contained in the TCP RCV.WND field will
RLWND contained in the RCV.WND field of TCP will be in general generally be smaller than the congestion window calculated by the TCP
smaller than the congestion window calculated by the TCP sender, sender, implying that the rLEDBAT congestion control algorithm will
implying that the rLEDBAT congestion control algorithm will be be effectively controlling the sender's window. One exception to
effectively controlling the sender's window. One exception to this this scenario is that at the beginning of the connection, when there
is at the beginning of the connection, when there is no information is no information to set RLWND, RLWND is set to its maximum value, so
to set RLWND, then, RLWND is set to its maximum value, so that the that the sending rate of the sender is governed by the flow control
sending rate of the sender is governed by the flow control algorithm algorithm of the receiver and the TCP slow start mechanism of the
of the receiver and the TCP slow start mechanism of the sender. sender.
In summary, the sender's window is: SND.WND = min(cwnd, RLWND, fcwnd) In summary, the sender's window is SND.WND = min(cwnd, RLWND, fcwnd)
3.1.1. Avoiding window shrinking 4.1.1. Avoiding Window Shrinking
The LEDBAT/LEDBAT++ algorithm executed in a rLEDBAT receiver The LEDBAT/LEDBAT++ algorithm executed in a rLEDBAT receiver
increases or decreases RLWND according to congestion signals increases or decreases RLWND according to congestion signals
(variations on the estimated queueing delay and packet loss). If (variations in the estimated queuing delay and packet loss). If
RLWND is decreased and directly announced in RCV.WND, this could lead RLWND is decreased and directly announced in RCV.WND, this could lead
to an announced window that is smaller than what is currently in use. to an announced window that is smaller than what is currently in use.
This so called 'shrinking the window' is discouraged as per This so-called "shrinking the window" is discouraged as per
[RFC9293], as it may cause unnecessary packet loss and performance [RFC9293], as it may cause unnecessary packet loss and performance
penalty. To be consistent with [RFC9293], the rLEDBAT receiver penalties. To be consistent with [RFC9293], the rLEDBAT receiver
SHOULD NOT shrink the receive window. SHOULD NOT shrink the receive window.
In order to avoid window shrinking, the receiver MUST only reduce In order to avoid window shrinking, the receiver MUST only reduce
RCV.WND by the number of bytes upon of a received data packet. This RCV.WND by the number of bytes contained in a received data packet.
may fall short to honor the new calculated value of the RLWND This may fall short to honor the new calculated value of the RLWND
immediately. However, the receiver SHOULD progressively reduce the immediately. However, the receiver SHOULD progressively reduce the
advertised RCV.WND, always honoring that the reduction is less or advertised RCV.WND, always honoring that the reduction is less than
equal than the received bytes, until the target window determined by or equal to the received bytes, until the target window determined by
the rLEDBAT algorithm is reached. This implies that it may take up the rLEDBAT algorithm is reached. This implies that it may take up
to one RTT for the rLEDBAT receiver to drain enough in-flight bytes to one RTT for the rLEDBAT receiver to drain enough in-flight bytes
to completely close its receive window without shrinking it. This is to completely close its receive window without shrinking it. This is
sufficient to honor the window output from the LEDBAT/LEDBAT++ sufficient to honor the window output from the LEDBAT/LEDBAT++
algorithms since they only allow to perform at most one algorithms, since they are only allowed to perform at most one
multiplicative decrease per RTT. multiplicative decrease per RTT.
3.1.2. Setting the Window Scale Option 4.1.2. Setting the Window Scale Option
The Window Scale (WS) option [RFC7323] is a means to increase the The Window Scale (WS) option [RFC7323] is a means to increase the
maximum window size permitted by the Receive Window. The WS option maximum window size permitted by the receive window. The WS option
defines a scale factor which restricts the granularity of the receive defines a scale factor that restricts the granularity of the receive
window that can be announced. This means that the rLEDBAT client window that can be announced. This means that the rLEDBAT client
will have to accumulate the increases resulting from multiple will have to accumulate the increases resulting from multiple
received packets, and only convey a change in the window when the received packets and only convey a change in the window when the
accumulated sum of increases is equal or higher than one increase accumulated sum of increases is equal to or higher than one increase
step as imposed by the scaling factor according to the WS option in step as imposed by the scaling factor according to the WS option in
place for the TCP connection. place for the TCP connection.
Changes in the receive window that are smaller than 1 MSS are Changes in the receive window that are smaller than 1 MSS (Maximum
unlikely to have any immediate impact on the sender's rate, as usual Segment Size) are unlikely to have any immediate impact on the
TCP's segmentation practice results in sending full segments (i.e., sender's rate. As usual, TCP's segmentation practice results in
segments of size equal to the MSS). Current WS option specification sending full segments (i.e., segments of size equal to the MSS).
[RFC7323] defines that allowed values for the WS option are between 0 [RFC7323], which defines the WS option, specifies that allowed values
and 14. Assuming a MSS around 1500 bytes, WS option values between 0 for the WS option are between 0 and 14. Assuming an MSS of around
and 11 result in the receive window being expressed in units that are 1500 bytes, WS option values between 0 and 11 result in the receive
about 1 MSS or smaller. So, WS option values between 0 and 11 have window being expressed in units that are about 1 MSS or smaller. So,
no impact in rLEDBAT (unless packets smaller than the MSS are being WS option values between 0 and 11 have no impact in rLEDBAT (unless
exchanged). packets smaller than the MSS are being exchanged).
WS option values higher than 11 can affect the dynamics of rLEDBAT, WS option values higher than 11 can affect the dynamics of rLEDBAT,
since control may become too coarse (e.g., with WS of 14, a change in since control may become too coarse (e.g., with a WS option value of
one unit of the receive window implies a change of 10 MSS in the 14, a change in one unit of the receive window implies a change of 10
effective window). MSS in the effective window).
For the above reasons, the rLEDBAT client SHOULD set WS option values For the above reasons, the rLEDBAT client SHOULD set WS option values
lower than 12. Additional experimentation is required to explore the lower than 12. Additional experimentation is required to explore the
impact of larger WS values on rLEDBAT dynamics. impact of larger WS values on rLEDBAT dynamics.
Note that the recommendation for rLEDBAT to set the WS option value Note that the recommendation for rLEDBAT to set the WS option values
to lower values does not precludes the communication with servers to lower values does not preclude communication with servers that set
that set the WS option values to larger values, since the WS option the WS option values to larger values, since WS option values are set
value is set independently for each direction of the TCP connection. independently for each direction of the TCP connection.
3.2. Measuring delays 4.2. Measuring Delays
Both LEDBAT and LEDBAT++ measure base and current delays to estimate Both LEDBAT and LEDBAT++ measure base and current delays to estimate
the queueing delay. LEDBAT uses the one way delay while LEDBAT++ the queuing delay. LEDBAT uses the one-way delay, while LEDBAT++
uses the round trip time. In the next sections we describe how uses the RTT. In the next sections, we describe how rLEDBAT
rLEDBAT mechanisms enable the receiver to measure the one way delay mechanisms enable the receiver to measure the one-way delay or the
or the round trip time, whatever is needed depending on the RTT -- whichever is needed, depending on the congestion control
congestion control algorithm used. algorithm used.
3.2.1. Measuring RTT to estimate the queueing delay 4.2.1. Measuring RTT to Estimate the Queuing Delay
LEDBAT++ uses the round trip time (RTT) to estimate the queueing LEDBAT++ uses the RTT to estimate the queuing delay. In order to
delay. In order to estimate the queueing delay using RTT, the estimate the queuing delay using RTT, the rLEDBAT receiver estimates
rLEDBAT receiver estimates the base RTT (i.e., the constant the base RTT (i.e., the constant components of RTT) and also measures
components of RTT) and also measures the current RTT. By subtracting the current RTT. By subtracting these two values, we obtain the
these two values, we obtain the queuing delay to be used by the queuing delay to be used by the rLEDBAT controller.
rLEDBAT controller.
LEDBAT++ discovers the base RTT (RTTb) by taking the minimum value of LEDBAT++ discovers the base RTT (RTTb) by taking the minimum value of
the measured RTTs over a period of time. The current RTT (RTTc) is the measured RTTs over a period of time. The current RTT (RTTc) is
estimated using a number of recent samples and applying a filter, estimated using a number of recent samples and applying a filter,
such as the minimum (or the mean) of the last k samples. Using RTT such as the minimum (or the mean) of the last k samples. Using RTT
to estimate the queueing delay has a number of shortcomings and to estimate the queuing delay has a number of shortcomings and
difficulties that we discuss next. difficulties, as discussed below.
The queuing delay measured using RTT includes also the queueing delay The queuing delay measured using RTT also includes the queuing delay
experienced by the return packets in the direction from the rLEDBAT experienced by the return packets in the direction from the rLEDBAT
receiver to the sender. This is a fundamental limitation of this receiver to the sender. This is a fundamental limitation of this
approach. The impact of this error is that the rLEDBAT controller approach. The impact of this limitation is that the rLEDBAT
will also react to congestion in the reverse path direction which controller will also react to congestion in the reverse path
results in an even more conservative mechanism. direction, resulting in an even more conservative mechanism.
In order to measure RTT, the rLEDBAT client MUST enable the Time In order to measure RTT, the rLEDBAT client MUST enable the TS option
Stamp (TS) option [RFC7323]. By matching the TSVal value carried in [RFC7323]. By matching the TSval carried in outgoing packets with
outgoing packets with the TSecr value observed in incoming packets, the Timestamp Echo Reply (TSecr) value [RFC7323] observed in incoming
it is possible to measure RTT. This allows the rLEDBAT receiver to packets, it is possible to measure RTT. This allows the rLEDBAT
measure RTT even if it is acting as a pure receiver. In a pure receiver to measure RTT even if it is acting as a pure receiver. In
receiver there is no data flowing from the rLEDBAT receiver to the a pure receiver, there is no data flowing from the rLEDBAT receiver
sender, making impossible to match data packets with acknowledgements to the sender, making it impossible to match data packets with
packets to measure RTT, as it is usually done in TCP for other Acknowledgment packets to measure RTT, in contrast to what is usually
purposes. done in TCP for other purposes.
Depending on the frequency of the local clock used to generate the Depending on the frequency of the local clock used to generate the
values included in the TS option, several packets may carry the same values included in the TS option, several packets may carry the same
TSVal value. If that happens, the rLEDBAT receiver will be unable to TSval. If that happens, the rLEDBAT receiver will be unable to match
match the different outgoing packets carrying the same TSVal value the different outgoing packets carrying the same TSval with the
with the different incoming packets carrying also the same TSecr different incoming packets also carrying the same TSecr value.
value. However, it is not necessary for rLEDBAT to use all packets However, it is not necessary for rLEDBAT to use all packets to
to estimate RTT and sampling a subset of in-flight packets per RTT is estimate RTT, and sampling a subset of in-flight packets per RTT is
enough to properly assess the queueing delay. RTT MUST then be enough to properly assess the queuing delay. RTT MUST then be
calculated as the time since the first packet with a given TSVal was calculated as the time since the first packet with a given TSval was
sent and the first packet that was received with the same value sent and the first packet that was received with the same value
contained in the TSecr. Other packets with repeated TS values SHOULD contained in the TSecr. Other packets with repeated TS values SHOULD
NOT be used for RTT calculation. NOT be used for RTT calculations.
Several issues must be addressed in order to avoid an artificial Several issues must be addressed in order to avoid an artificial
increase of the observed RTT. Different issues emerge depending increase in the observed RTT. Different issues emerge, depending on
whether the rLEDBAT capable host is sending data packets or pure ACKs whether the rLEDBAT-capable host is sending data packets or pure ACKs
to measure RTT. We next consider the issues separately. to measure RTT. We next consider these issues separately.
3.2.1.1. Measuring RTT sending pure ACKs 4.2.1.1. Measuring RTT When Sending Pure ACKs
In this scenario, the rLEDBAT node (node A) sends a pure ACK to the In this scenario, the rLEDBAT node (node A) sends a pure ACK to the
other endpoint of the TCP connection (node B), including the TS other endpoint of the TCP connection (node B), including the TS
option. Upon the reception of the TS Option, host B will copy the option. Upon the reception of the TS option, host B will copy the
value of the TSVal into the TSecr field of the TS option and include value of the TSval into the TSecr field of the TS option and include
that option into the next data packet towards host A. However, there that option in the next data packet towards host A. However, there
are two reasons why B may not send a packet immediately back to A, are two reasons why B may not send a packet immediately back to A,
artificially increasing the measured RTT. The first reason is when A artificially increasing the measured RTT. The first reason is when A
has no data to send. The second is when A has no available window to has no data to send. The second is when A has no available window to
put more packets in-flight. We describe next how each of these cases put more packets in flight. We next describe how each of these cases
is addressed. is addressed.
The case where the host B has no data to send when it receives the The case where host B has no data to send when it receives the pure
pure Acknowledgement is expected to be rare in the rLEDBAT use cases. Acknowledgment is expected to be rare in the rLEDBAT use
rLEDBAT will be used mostly for background file transfers so the cases. rLEDBAT will be used mostly for background file transfers, so
expected common case is that the sender will have data to send the expected common case is that the sender will have data to send
throughout the lifetime of the communication. However, if, for throughout the lifetime of the communication. However, if, for
example, the file is structured in blocks of data, it may be the case example, the file is structured in blocks of data, it may be the case
that the sender seldomly will have to wait until the next block is that the sender will seldom have to wait until the next block is
available to proceed with the data transfer. To address this available to proceed with the data transfer. To address this
situation, the filter used by the congestion control algorithm situation, the filter used by the congestion control algorithm
executed in the receiver SHOULD discard outliers (e.g. a min filter executed in the receiver SHOULD discard outliers (e.g., a MIN filter
would achieve this) when measuring RTT using pure ACK packets. [RFC6817] would achieve this) when measuring RTT using pure ACK
packets.
This limitation of the sender's window can come either from the TCP This limitation of the sender's window can come from either the TCP
congestion window in host B or from the announced receive window from congestion window in host B or the announced receive window from
the rLEDBAT in host A. Normally, the receive window will be the one rLEDBAT in host A. Normally, the receive window will be the one to
to limit the sender's transmission rate, since the LBE congestion limit the sender's transmission rate, since the LBE congestion
control algorithm used by the rLEDBAT node is designed to be more control algorithm used by the rLEDBAT node is designed to be more
restrictive on the sender's rate than standard-TCP. If the limiting restrictive on the sender's rate than standard-TCP. If the limiting
factor is the congestion window in the sender, it is less relevant if factor is the congestion window in the sender, it is less relevant if
rLEDBAT further reduces the receive window due to a bloated RTT rLEDBAT further reduces the receive window due to a bloated RTT
measurement, since the rLEDBAT node is not actively controlling the measurement, since the rLEDBAT node is not actively controlling the
sender's rate. Nevertheless, the proposed approach to discard larger sender's rate. Nevertheless, the proposed approach to discard larger
samples would also address this issue. samples would also address this issue.
To address the case in which the limiting factor is the receive To address the case in which the limiting factor is the receive
window announced by rLEDBAT, the congestion control algorithm at the window announced by rLEDBAT, the congestion control algorithm at the
receiver SHOULD discard RTT measurements during the window reduction receiver SHOULD discard RTT measurements during the window reduction
phase that are triggered by pure ACK packets. The rLEDBAT receiver phase that are triggered by pure ACK packets. The rLEDBAT receiver
is aware whether a given TSVal value was sent in a pure ACK packet is aware of whether a given TSval was sent in a pure ACK packet where
where the window was reduced, and if so, it can discard the the window was reduced, and if so, it can discard the corresponding
corresponding RTT measurement. RTT measurement.
3.2.1.2. Measuring RTT when sending data packets 4.2.1.2. Measuring RTT When Sending Data Packets
In the case that the rLEDBAT node is sending data packets and In the case that the rLEDBAT node is sending data packets and
matching them with pure ACKs to measure RTT, a factor that can matching them with pure ACKs to measure RTT, a factor that can
artificially increase the RTT measured is the presence of delayed artificially increase the RTT measured is the presence of delayed
Acknowledgements. According to the TS option generation rules Acknowledgments. According to the TS option generation rules
[RFC7323], the value included in the TSecr for a delayed ACK is the [RFC7323], the value included in the TSecr for a delayed ACK is the
one in the TSVal field of the earliest unacknowledged segment. This one in the TSval field of the earliest unacknowledged segment. This
may artificially increase the measured RTT. may artificially increase the measured RTT.
If both endpoints of the connection are sending data packets, If both endpoints of the connection are sending data packets,
Acknowledgments are piggybacked into the data packets and they are Acknowledgments are piggybacked onto the data packets and they are
not delayed. Delayed ACKs only increase RTT measurements in the case not delayed. Delayed ACKs only increase RTT measurements in the case
that the sender has no data to send. Since the expected use case for that the sender has no data to send. Since the expected use case for
rLEDBAT is that the sender will be sending background traffic to the rLEDBAT is that the sender will be sending background traffic to the
rLEDBAT receiver, the cases where delayed ACKs increase the measured rLEDBAT receiver, the cases where delayed ACKs increase the measured
RTT are expected to be rare. RTT are expected to be rare.
Nevertheless, measurements based on data packets from the rLEDBAT Nevertheless, measurements based on data packets from the rLEDBAT
node matching pure ACKs from the other end will result in an node matching pure ACKs from the other end will result in an
increased RTT sample. The additional increase in the measured RTT increased RTT sample. The additional increase in the measured RTT
will be up to 500 ms. The reason for this is that delayed ACKs are will be up to 500 ms. This is because delayed ACKs are generated
generated every second data packet received and not delayed more than every second data packet received and not delayed more than 500 ms
500 ms according to [RFC9293]. The rLEDBAT receiver MAY discard RTT according to [RFC9293]. The rLEDBAT receiver MAY discard RTT
measurements done using data packets from the rLEBDAT receiver and measurements done using data packets from the rLEDBAT receiver and
matching pure ACKs, especially if it has recent measurements done matching pure ACKs, especially if it has recent measurements done
using other packet combinations. Also, applying a filter that using other packet combinations. Applying a filter (e.g., a MIN
discards outliers would also address this issue (e.g. a min filter). filter) that discards outliers would also address this issue.
3.2.2. Measuring one way delay to estimate the queueing delay 4.2.2. Measuring One-Way Delay to Estimate the Queuing Delay
The LEDBAT algorithm uses the one-way delay of packets as input. A The LEDBAT algorithm uses the one-way delay of packets as input. A
TCP receiver can measure the delay of incoming packets directly (as TCP receiver can measure the delay of incoming packets directly (as
opposed to the sender-based LEDBAT, where the receiver measures the opposed to the sender-based LEDBAT, where the receiver measures the
one-way delay and needs to convey it to the sender). one-way delay and needs to convey it to the sender).
In the case of TCP, the receiver can use the TimeStamp option to In the case of TCP, the receiver can use the TS option to measure the
measure the one way delay by subtracting the timestamp contained in one-way delay by subtracting the timestamp contained in the incoming
the incoming packet from the local time at which the packet has packet from the local time at which the packet has arrived. As noted
arrived. As noted in [RFC6817] the clock offset between the clock of in [RFC6817], the clock offset between the sender's clock and the
the sender and the clock in the receiver does not affect the LEDBAT receiver's clock does not affect the LEDBAT operation, since LEDBAT
operation, since LEDBAT uses the difference between the base one way uses the difference between the base one-way delay and the current
delay and the current one way delay to estimate the queuing delay, one-way delay to estimate the queuing delay, effectively "canceling
effectively canceling the clock offset error in the queueing delay out" the clock offset error in the queuing delay estimation. There
estimation. There are however two other issues that the rLEDBAT are, however, two other issues that the rLEDBAT receiver needs to
receiver needs to take into account in order to properly estimate the take into account in order to properly estimate the one-way delay,
one way delay, namely, the units in which the received timestamps are namely the units in which the received timestamps are expressed and
expressed and the clock skew. We address them next. the clock skew. These issues are addressed below.
In order to measure the one way delay using TCP timestamps, the In order to measure the one-way delay using TCP timestamps, the
rLEDBAT receiver, first, needs to discover the units of values in the rLEDBAT receiver first needs to discover the units of values in the
TS option and, second, needs to account for the skew between the two TS option and then needs to account for the skew between the two
endpoint clocks. Note that a mismatch of 100 ppm (parts per million) endpoint clocks. Note that a mismatch of 100 ppm (parts per million)
in the estimation of the sender's clock rate accounts for 6 ms of in the estimation of the sender's clock rate accounts for 6 ms of
variation per minute in the measured delay. This just one order of variation per minute in the measured delay. This is just one order
magnitude below the target delay set by rLEDBAT (or potentially more of magnitude below the target delay set by rLEDBAT (or potentially
if the target is set to lower values, which is possible). Typical more if the target is set to lower values, which is possible).
skew for untrained clocks is reported to be around 100-200 ppm Typical skew for untrained clocks is reported to be around 100-200
[RFC6817]. ppm [RFC6817].
In order to learn both the TS units and the clock skew, the rLEDBAT In order to learn both the TS units and the clock skew, the rLEDBAT
receiver measures how much local time has elapsed between two packets receiver measures how much local time has elapsed between two packets
with different TS values issued by the sender. By comparing the with different TS values issued by the sender. By comparing the
local time difference and the TS value difference, the receiver can local time difference and the TS value difference, the receiver can
assess the TS units and relative clock skews. In order for this to assess the TS units and relative clock skews. In order for this to
be accurate, the packets carrying the different TS values should be accurate, the packets carrying the different TS values should
experience equal (or at least similar delay) when traveling from the experience equal (or at least similar) delay when traveling from the
sender to the receiver, as any difference in the experienced delays sender to the receiver, as any difference in the experienced delays
would introduce error in the unit/skew estimation. One possible would introduce an error in the unit/skew estimation. One possible
approach is to select packets that experienced the minimum delay approach is to select packets that experienced minimal delay (i.e.,
(i.e. close to zero queueing delay) to make the estimations. queuing delay close to zero) to make the estimations.
An additional difficulty regarding the estimation of the TS units and An additional difficulty regarding the estimation of the TS units and
clock skew in the context of (r)LEDBAT is that the LEDBAT congestion clock skew in the context of (r)LEDBAT is that the LEDBAT congestion
controller actions directly affect the (queueing) delay experienced controller actions directly affect the (queuing) delay experienced by
by packets. In particular, if there is an error in the estimation of packets. In particular, if there is an error in the estimation of
the TS units/skew, the LEDBAT controller will attempt to compensate the TS units/skew, the LEDBAT controller will attempt to compensate
it by reducing/increasing the load. The result is that the LEDBAT for it by reducing/increasing the load. The result is that the
operation interferes with the TS units/clock skew measurements. LEDBAT operation interferes with the TS units/clock skew
Because of this, measurements are more accurate when there is no measurements. Because of this, measurements are more accurate when
traffic in the connection (in addition to the packets used for the there is no traffic in the connection (in addition to the packets
measurements). The problem is that the receiver is unaware if the used for the measurements). The problem is that the receiver is
sender is injecting traffic at any point in time, and so, it is unaware of whether the sender is injecting traffic at any point in
unable to use these quiet intervals to perform measurements. The time; it is therefore unable to use these quiet intervals to perform
receiver can however, force periodic slowdowns, reducing the measurements. The receiver can, however, force periodic slowdowns,
announced receive window to a few packets and perform the reducing the announced receive window to a few packets and performing
measurements then. the measurements at that time.
It is possible for the rLEDBAT receiver to perform multiple It is possible for the rLEDBAT receiver to perform multiple
measurements to assess both the TS units and the relative clock skew measurements to assess both the TS units and the relative clock skew
during the lifetime of the connection, in order to obtain more during the lifetime of the connection, in order to obtain more
accurate results. Clock skew measurements are more accurate if the accurate results. Clock skew measurements are more accurate if the
time period used to discover the skew is larger, as the impact of the time period used to discover the skew is larger, as the impact of the
skew becomes more apparent. It is a reasonable approach for the skew becomes more apparent. It is a reasonable approach for the
rLEDBAT receiver to perform an early discovery of the TS units (and rLEDBAT receiver to perform an early discovery of the TS units (and
the clock skew) using the first few packets of the TCP connection and the clock skew) using the first few packets of the TCP connection and
then improve the accuracy of the TS units/clock skew estimation using then improve the accuracy of the TS units/clock skew estimation using
periodic measurements later in the lifetime of the connection. periodic measurements later in the lifetime of the connection.
3.3. Detecting packet losses and retransmissions 4.3. Detecting Packet Losses and Retransmissions
The rLEDBAT receiver is capable of detecting retransmitted packets in The rLEDBAT receiver is capable of detecting retransmitted packets as
the following way. We call RCV.HGH the highest sequence number follows. We call RCV.HGH the highest sequence number corresponding
corresponding to a received byte of data (not assuming that all bytes to a received byte of data (not assuming that all bytes with smaller
with smaller sequence numbers have been received already, there may sequence numbers have been received already, there may be holes), and
be holes) and we call TSV.HGH the TSVal value corresponding to the we call TSV.HGH the TSval corresponding to the segment in which that
segment in which that byte was carried. SEG.SEQ stands for the byte was carried. SEG.SEQ stands for the sequence number of a newly
sequence number of a newly received segment and we call TSV.SEQ the received segment, and we call TSV.SEQ the TSval of the newly received
TSVal value of the newly received segment. segment.
If SEG.SEQ < RCV.HGH and TSV.SEQ > TSV.HGH then the newly received If SEG.SEQ < RCV.HGH and TSV.SEQ > TSV.HGH, then the newly received
segment is a retransmission. This is so because the newly received segment is a retransmission. This is so because the newly received
segment was generated later than another already received segment segment was generated later than another already-received segment
which contained data with a larger sequence number. This means that that contained data with a larger sequence number. This means that
this segment was lost and was retransmitted. this segment was lost and was retransmitted.
The proposed mechanism to detect retransmissions at the receiver The proposed mechanism to detect retransmissions at the receiver
fails when there are window tail drops. If all packets in the tail fails when there are window tail drops. If all packets in the tail
of the window are lost, the receiver will not be able to detect a of the window are lost, the receiver will not be able to detect a
mismatch between the sequence numbers of the packets and the order of mismatch between the sequence numbers of the packets and the order of
the timestamps. In this case, rLEDBAT will not react to losses but the timestamps. In this case, rLEDBAT will not react to losses;
the TCP congestion controller at the sender will, most likely however, the TCP congestion controller at the sender will, most
reducing its window to 1MSS and take over the control of the sending likely reducing its window to 1 MSS and taking over the control of
rate, until slow start ramps up and catches the current value of the the sending rate until slow start ramps up and catches the current
rLEDBAT window. value of the rLEDBAT window.
4. Experiment Considerations 5. Experiment Considerations
The status of this document is Experimental. The general purpose of The status of this document is Experimental. The general purpose of
the proposed experiment is to gain more experience running rLEDBAT the proposed experiment is to gain more experience running rLEDBAT
over different network paths to see if the proposed rLEDBAT over different network paths to see if the proposed rLEDBAT
parameters perform well in different situations. Specifically, we parameters perform well in different situations. Specifically, we
would like to learn about the following aspects of the rLEDBAT would like to learn about the following aspects of the rLEDBAT
mechanism: mechanism:
- Interaction between the sender and the receiver Congestion * Interaction between the sender's and receiver's congestion control
control algorithms. rLEDBAT posits that because the rLEDBAT algorithms. rLEDBAT posits that because the rLEDBAT receiver is
receiver is using a less-than-best-effort congestion control using a less-than-best-effort congestion control algorithm, the
algorithm, the receiver congestion control algorithm will expose a receiver's congestion control algorithm will expose a smaller
smaller congestion window (conveyed though the Receive Window) congestion window (conveyed through the receive window) than the
than the one resulting from the congestion control algorithm one resulting from the congestion control algorithm executed at
executed at the sender. One of the purposes of the experiment is the sender. One of the purposes of the experiment is to learn how
learn how these two interact and if the assumption that the these two algorithms interact and if the assumption that the
receiver side is always controlling the sender's rate (and making receiver side is always controlling the sender's rate (and making
rLEDBAT effective) holds. The experiment should include the rLEDBAT effective) holds. The experiment should include the
different congestion control algorithms that are currently widely different congestion control algorithms that are currently widely
used in the Internet, including Cubic, BBR and LEDBAT(++). used in the Internet, including CUBIC, Bottleneck Bandwidth and
Round-trip propagation time (BBR), and LEDBAT(++).
- Interaction between rLEDBAT and Active Queue Management * Interaction between rLEDBAT and Active Queue Management techniques
techniques such as Codel, PIE and L4S. such as Controlled Delay (CoDel); Proportional Integral controller
Enhanced (PIE); and Low Latency, Low Loss, and Scalable Throughput
(L4S).
- How the rLEDBAT should resume after a period during which there * How rLEDBAT should resume after a period during which there was no
was no incoming traffic and the information about the rLEDBAT incoming traffic and the information about the rLEDBAT state
state information is potentially dated. information is potentially dated.
4.1. Status of the experiment at the time of this writing. 5.1. Status of the Experiment at the Time of This Writing
Currently there are the following implementations of rLEDBAT that can Currently, the following implementations of rLEDBAT can be used for
be used for experimentation: experimentation:
- Windows 11. rLEDBAT is available in Microsoft's Windows 11 22H2 * Windows 11. rLEDBAT is available in Microsoft's Windows 11 22H2
since October 2023 [Windows11]. since October 2023 [Windows11].
- Windows Server 2022. rLEDBAT is available in Microsoft's Windows * Windows Server 2022. rLEDBAT is available in Microsoft's Windows
Server 2022 since September 2022 [WindowsServer]. Server 2022 since September 2022 [WindowsServer].
- Apple. rLEDBAT is available in MacOS and iOS since 2021 [Apple]. * Apple. rLEDBAT is available in macOS and iOS since 2021 [Apple].
- Linux implementation, open source, available since 2022 at * Linux implementation, open source, available since 2022
https://github.com/net-research/rledbat_module. [rledbat_module].
- ns3 implementation, open source, available since 2020 at * ns3 implementation, open source, available since 2020
https://github.com/manas11/implementation-of-rLEDBAT-in-ns-3. [rLEDBAT-in-ns-3].
In addition, rLEDBAT has been deployed by Microsoft in wide scale in In addition, rLEDBAT has been deployed by Microsoft at wide scale in
the following services: the following services:
- BITS (Background Intelligent Transfer Service) * BITS (Background Intelligent Transfer Service)
- DO (Delivery Optimization) service * DO (Delivery Optimization) service
- Windows update # using DO * Windows update: using DO
- Windows Store # using DO
- OneDrive * Windows Store: using DO
- Windows Error Reporting # wermgr.exe; werfault.exe * OneDrive
- System Center Configuration Manager (SCCM) * Windows Error Reporting: wermgr.exe; werfault.exe
- Windows Media Player * System Center Configuration Manager (SCCM)
- Microsoft Office * Windows Media Player
- Xbox (download games) # using DO * Microsoft Office
* Xbox (download games): using DO
Some initial experiments involving rLEDBAT have been reported in Some initial experiments involving rLEDBAT have been reported in
[COMNET3]. Experiments involving the interaction of LEDBAT++ and BBR [COMNET3]. Experiments involving the interaction between LEDBAT++
are presented in [COMNET2]. An experimental evaluation of the and BBR are presented in [COMNET2]. An experimental evaluation of
LEDBAT++ algorithm is presented in [COMNET1]. As LEDBAT++ is one of the LEDBAT++ algorithm is presented in [COMNET1]. As LEDBAT++ is one
the less-than-best-effort congestion control algorithms that rLEDBAT of the less-than-best-effort congestion control algorithms that
relies on, the results regarding LEDBAT++ interaction with other rLEDBAT relies on, the results regarding how LEDBAT++ interacts with
congestion control algorithms are relevant for the understanding of other congestion control algorithms are relevant for the
rLEDBAT as well. understanding of rLEDBAT as well.
5. Security Considerations 6. Security Considerations
Overall, we believe that rLEDBAT does not introduce any new Overall, we believe that rLEDBAT does not introduce any new
vulnerabilities to existing TCP endpoints, as it relies on existing vulnerabilities to existing TCP endpoints, as it relies on existing
TCP knobs, notably the Receive Window and timestamps. TCP knobs, notably the receive window and timestamps.
Specifically, rLEDBAT uses RCV.WND to modulate the rate of the Specifically, rLEDBAT uses RCV.WND to modulate the rate of the
sender. An attacker wishing to starve a flow can simply reduce the sender. An attacker wishing to starve a flow can simply reduce the
RCV.WND, irrespective of whether rLEDBAT is being used or not. RCV.WND, irrespective of whether rLEDBAT is being used or not.
We can further ask ourselves whether the attacker can use the rLEDBAT We can further ask ourselves whether the attacker can use the rLEDBAT
mechanisms in place to force the rLEDBAT receiver to reduce the RCV mechanisms in place to force the rLEDBAT receiver to reduce the
WND. There are two ways an attacker can do that. One would be to RCV.WND. There are two ways an attacker can do this:
introduce an artificial delay to the packets either by actually
delaying the packets or modifying the Timestamps. This would cause
the rLEDBAT receiver to believe that a queue is building up and
reduce the RCV.WND. Note that an attacker to do that must be on
path, so if that is the case, it is probably more direct to simply
reduce the RCV.WND.
The other option would be for the attacker to make the rLEDBAT * One would be to introduce an artificial delay to the packets by
receiver believe that a loss has occurred. To do that, it basically either actually delaying the packets or modifying the timestamps.
needs to retransmit an old packet (to be precise, it needs to This would cause the rLEDBAT receiver to believe that a queue is
transmit a packet with the right sequence number and the right port building up and reduce the RCV.WND. Note that to do so, an
and IP numbers). This means that the attacker can achieve a attacker must be on path, so if that is the case, it is probably
reduction of incoming traffic to the rLEDBAT receiver not only by more direct to simply reduce the RCV.WND.
modifying the RCV.WND field of the packets originated from the
rLEDBAT host, but also by injecting packets with the proper sequence
number in the other direction. This may slightly expand the attack
surface.
6. IANA Considerations * The other option would be for the attacker to make the rLEDBAT
receiver believe that a loss has occurred. To do this, it
basically needs to retransmit an old packet (to be precise, it
needs to transmit a packet with the correct sequence number and
the correct port and IP numbers). This means that the attacker
can achieve a reduction of incoming traffic to the rLEDBAT
receiver not only by modifying the RCV.WND field of the packets
originated from the rLEDBAT host but also by injecting packets
with the proper sequence number in the other direction. This may
slightly expand the attack surface.
No actions are required from IANA. 7. IANA Considerations
7. Acknowledgements This document has no IANA actions.
This work was supported by the EU through the StandICT projects RXQ, 8. References
CCI and CEL6, the NGI Pointer RIM project and the H2020 5G-RANGE
project and by the Spanish Ministry of Economy and Competitiveness
through the 5G-City project (TEC2016-76795-C6-3-R).
We would like to thank ICCRG chairs Reese Enghardt and Vidhi Goel for 8.1. Normative References
their support on this work. We would also like to thank Daniel Havey
for his help. We would like to thank Colin Perkins, Mirja
Kuehlewind, and Vidhi Goel for their reviews and comments on earlier
versions of this document.
8. Informative References [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
[Apple] Stuart, S.C. and V.G. Vidhi, "Reduce network delays for [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
your app", WWDC21 https://developer.apple.com/videos/play/ 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
wwdc2021/10239/, 2021. May 2017, <https://www.rfc-editor.org/info/rfc8174>.
[COMNET1] Bagnulo, M.B. and A.G. Garcia-Martinez, "An experimental 8.2. Informative References
evaluation of LEDBAT++", Computer Networks Volume 212,
2022.
[COMNET2] Bagnulo, M.B. and A.G. Garcia-Martinez, "When less is [Apple] Cheshire, S. and V. Goel, "Reduce network delays for your
more: BBR versus LEDBAT++", Computer Networks Volume 219, app", Apple Worldwide Developers Conference (WWDC2021),
2022. Video, 2021,
<https://developer.apple.com/videos/play/wwdc2021/10239/>.
[COMNET3] Bagnulo, M.B., Garcia-Martinez, A.G., Mandalari, A.M., [COMNET1] Bagnulo, M. and A. García-Martínez, "An experimental
Balasubramanian, P.B,., Havey, D.H., and G.M. Montenegro, evaluation of LEDBAT++", Computer Networks, vol. 212,
DOI 10.1016/j.comnet.2022.109036, July 2022,
<https://doi.org/10.1016/j.comnet.2022.109036>.
[COMNET2] Bagnulo, M. and A. García-Martínez, "When less is more:
BBR versus LEDBAT++", Computer Networks, vol. 219,
DOI 10.1016/j.comnet.2022.109460, December 2022,
<https://doi.org/10.1016/j.comnet.2022.109460>.
[COMNET3] Bagnulo, M., García-Martínez, A., Mandalari, A.M.,
Balasubramanian, P., Havey, D., and G. Montenegro,
"Design, implementation and validation of a receiver- "Design, implementation and validation of a receiver-
driven less-than-best-effort transport", Computer driven less-than-best-effort transport", Computer
Networks Volume 233, 2022. Networks, vol. 233, DOI 10.1016/j.comnet.2023.109841,
September 2023,
<https://doi.org/10.1016/j.comnet.2023.109841>.
[I-D.irtf-iccrg-ledbat-plus-plus] [LEDBAT++] Balasubramanian, P., Ertugay, O., Havey, D., and M.
Balasubramanian, P., Ertugay, O., and D. Havey, "LEDBAT++: Bagnulo, "LEDBAT++: Congestion Control for Background
Congestion Control for Background Traffic", Work in Traffic", Work in Progress, Internet-Draft, draft-irtf-
Progress, Internet-Draft, draft-irtf-iccrg-ledbat-plus- iccrg-ledbat-plus-plus-03, 9 September 2025,
plus-01, 25 August 2020,
<https://datatracker.ietf.org/doc/html/draft-irtf-iccrg- <https://datatracker.ietf.org/doc/html/draft-irtf-iccrg-
ledbat-plus-plus-01>. ledbat-plus-plus-03>.
[RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion [RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion
Control", RFC 5681, DOI 10.17487/RFC5681, September 2009, Control", RFC 5681, DOI 10.17487/RFC5681, September 2009,
<https://www.rfc-editor.org/info/rfc5681>. <https://www.rfc-editor.org/info/rfc5681>.
[RFC6582] Henderson, T., Floyd, S., Gurtov, A., and Y. Nishida, "The
NewReno Modification to TCP's Fast Recovery Algorithm",
RFC 6582, DOI 10.17487/RFC6582, April 2012,
<https://www.rfc-editor.org/info/rfc6582>.
[RFC6817] Shalunov, S., Hazel, G., Iyengar, J., and M. Kuehlewind, [RFC6817] Shalunov, S., Hazel, G., Iyengar, J., and M. Kuehlewind,
"Low Extra Delay Background Transport (LEDBAT)", RFC 6817, "Low Extra Delay Background Transport (LEDBAT)", RFC 6817,
DOI 10.17487/RFC6817, December 2012, DOI 10.17487/RFC6817, December 2012,
<https://www.rfc-editor.org/info/rfc6817>. <https://www.rfc-editor.org/info/rfc6817>.
[RFC7323] Borman, D., Braden, B., Jacobson, V., and R. [RFC7323] Borman, D., Braden, B., Jacobson, V., and R.
Scheffenegger, Ed., "TCP Extensions for High Performance", Scheffenegger, Ed., "TCP Extensions for High Performance",
RFC 7323, DOI 10.17487/RFC7323, September 2014, RFC 7323, DOI 10.17487/RFC7323, September 2014,
<https://www.rfc-editor.org/info/rfc7323>. <https://www.rfc-editor.org/info/rfc7323>.
[RFC9293] Eddy, W., Ed., "Transmission Control Protocol (TCP)", [RFC9293] Eddy, W., Ed., "Transmission Control Protocol (TCP)",
STD 7, RFC 9293, DOI 10.17487/RFC9293, August 2022, STD 7, RFC 9293, DOI 10.17487/RFC9293, August 2022,
<https://www.rfc-editor.org/info/rfc9293>. <https://www.rfc-editor.org/info/rfc9293>.
[RFC9438] Xu, L., Ha, S., Rhee, I., Goel, V., and L. Eggert, Ed., [RFC9438] Xu, L., Ha, S., Rhee, I., Goel, V., and L. Eggert, Ed.,
"CUBIC for Fast and Long-Distance Networks", RFC 9438, "CUBIC for Fast and Long-Distance Networks", RFC 9438,
DOI 10.17487/RFC9438, August 2023, DOI 10.17487/RFC9438, August 2023,
<https://www.rfc-editor.org/info/rfc9438>. <https://www.rfc-editor.org/info/rfc9438>.
[Windows11] [rLEDBAT-in-ns-3]
Forsmann, C.F., "What's new in Delivery Optimization", "Implementation-of-rLEDBAT-in-ns-3", commit 2ab34ad, 24
Microsoft Documentation https://learn.microsoft.com/en- June 2020, <https://github.com/manas11/implementation-of-
us/windows/deployment/do/whats-new-do, 2023. rLEDBAT-in-ns-3>.
[WindowsServer]
Havey, D.H., "LEDBAT Background Data Transfer for
Windows", Microsoft Blog
https://techcommunity.microsoft.com/t5/networking-
blog/ledbat-background-data-transfer-for-windows/ba-
p/3639278, 2022.
Appendix A. Terminology
We use the following abreviations thoughout the text. We include a
short list for the reader's convenence:
RCV.WND: the value included in the Receive Window field of the TCP
header (which computation is modified by this specification)
SND.WND: The TCP sender's window
cwnd: the consgestion window as computed by the congestion control
algorithm running at the TCP sender.
RLWND: the window value calculated by rLEDBAT algorithm
fcwnd: the value that a standard RFC793bis TCP receiver calculates
to set in the receive window for flow control purposes.
RCV.HGH: the highest sequence number corresponding to a received
byte of data at one point in time
TSV.HGH: TSV.HGH the TSVal value corresponding to the segment in [rledbat_module]
which RCV.HGH was carried at that point in time "rledbat_module", commit d82ff20, 9 September 2022,
<https://github.com/net-research/rledbat_module>.
SEG.SEQ: the sequence number of the last received segment [Windows11]
Microsoft, "What's new in Delivery Optimization",
Microsoft Windows Documentation, October 2024,
<https://learn.microsoft.com/en-us/windows/deployment/do/
whats-new-do>.
TSV.SEQ: the TSVal value of the last received segment [WindowsServer]
Havey, D., "LEDBAT Background Data Transfer for Windows",
Microsoft Networking Blog, September 2022,
<https://techcommunity.microsoft.com/t5/networking-blog/
ledbat-background-data-transfer-for-windows/ba-p/3639278>.
Appendix B. rLEDBAT pseudo-code Appendix A. rLEDBAT Pseudocode
We next describe how to integrate the proposed rLEDBAT mechanisms and In this section, we describe how to integrate the proposed rLEDBAT
an LBE delay-based congestion control algorithm such as LEDBAT or mechanisms and an LBE delay-based congestion control algorithm such
LEDBAT++. We describe the integrated algorithm as two procedures, one as LEDBAT or LEDBAT++. We describe the integrated algorithm as two
that is executed when a packet is received by a rLEDBAT-enabled procedures: one that is executed when a packet is received by a
endpoint (Figure 2) and another that is executed when the rLEDBAT- rLEDBAT-enabled endpoint (Figure 2) and another that is executed when
enabled endpoint sends a packet (Figure 3). At the beginning, RLWND the rLEDBAT-enabled endpoint sends a packet (Figure 3). At the
is set to its maximum value, so that the sending rate of the sender beginning, RLWND is set to its maximum value, so that the sending
is governed by the flow control algorithm of the receiver and the TCP rate of the sender is governed by the flow control algorithm of the
slow start mechanism of the sender, and the ackedBytes variable is receiver and the TCP slow start mechanism of the sender, and the
set to 0. ackedBytes variable is set to 0.
We assume that the LBE congestion control algorithm defines a We assume that the LBE congestion control algorithm defines a
WindowIncrease() function and a WindowDecrease() function. For WindowIncrease() function and a WindowDecrease() function. For
example, in the case of LEDBAT++, the WindowIncrease() function is an example, in the case of LEDBAT++, the WindowIncrease() function is an
additive increase, while the WindowDecrease() function is a additive increase, while the WindowDecrease() function is a
multiplicative decrease. In the case of the WindowIncrease(), we multiplicative decrease. In the case of the WindowIncrease()
assume that it takes as input the current window size and the number function, we assume that it takes as input the current window size
of bytes that were acknowledged since the last window update and the number of bytes that were acknowledged since the last window
(ackedBytes) and returns as output the updated window size. In the update (ackedBytes) and returns as output the updated window size.
case of WindowDecrease(), it takes as input the current window size In the case of the WindowDecrease() function, it takes as input the
and returns the updated window size. current window size and returns the updated window size.
The data structures used in the algorithms are as follows. The The data structures used in the algorithms are as follows. The
sentList is a list that contains the TSval and the local send time of sendList is a list that contains the TSval and the local send time of
each packet sent by the rLEDBAT-enabled endpoint. The TSecr field of each packet sent by the rLEDBAT-enabled endpoint. The TSecr field of
the packets received by the rLEDBAT-enabled endpoint are matched with the packets received by the rLEDBAT-enabled endpoint is matched with
the sendList to compute the RTT. the sendList to compute the RTT.
The RTT values computed for each received packet are stored in the The RTT values computed for each received packet are stored in the
RTTlist, which contains also the received TSecr (to avoid using RTTlist, which also contains the received TSecr (to avoid using
multiple packets with the same TSecr for RTT calculations, only the multiple packets with the same TSecr for RTT calculations, only the
first packet received for a given TSecr is used to compute the RTT). first packet received for a given TSecr is used to compute the RTT).
It also contains the local time at which the packet was received, to It also contains the local time at which the packet was received, to
allow selecting the RTTs measured in a given period (e.g., in the allow selecting the RTTs measured in a given period (e.g., in the
last 10 minutes). RTTlist is initialized with all its values to its last 10 minutes). RTTlist is initialized with all its values to its
maximum. maximum.
procedure receivePacket() procedure receivePacket()
//Looks for first sent packet with same TSval as TSecr, and, //Looks for first sent packet with same TSval as TSecr, and
//returns time difference //returns time difference
receivedRTT = computeRTT(sentList, receivedTSecr, receivedTime) receivedRTT = computeRTT(sendList, receivedTSecr, receivedTime)
//Inserts minimum value for a given receivedTSecr //Inserts minimum value for a given receivedTSecr
//note that many received packets may contain same receivedTSecr //Note that many received packets may contain same receivedTSecr
insertRTT (RTTlist, receivedRTT, receivedTSecr, receivedTime) insertRTT (RTTlist, receivedRTT, receivedTSecr, receivedTime)
filteredRTT = minLastKMeasures(RTTlist, K=4) filteredRTT = minLastKMeasures(RTTlist, K=4)
baseRTT = minLastNSeconds(RTTlist, N=180) baseRTT = minLastNSeconds(RTTlist, N=180)
qd = filteredRTT - baseRTT qd = filteredRTT - baseRTT
//ackedBytes is the number of bytes that can be used to reduce //ackedBytes is the number of bytes that can be used to reduce
//the Receive Window - without shrinking it - if necessary //the receive window - without shrinking it - if necessary
ackedBytes = ackedBytes + receiveBytes ackedBytes = ackedBytes + receiveBytes
if retransmittedPacketDetected then if retransmittedPacketDetected then
RLWND = DecreaseWindow(RLWND) // Only once per RTT RLWND = DecreaseWindow(RLWND) //Only once per RTT
end if end if
if qd < T then if qd < T then
RLWND = IncreaseWindow(RLWND, ackedBytes) RLWND = IncreaseWindow(RLWND, ackedBytes)
else else
RLWND = DecreaseWindow(RLWND) RLWND = DecreaseWindow(RLWND)
end if end if
end procedure end procedure
Figure 2: Procedure executed when a packet is received Figure 2: Procedure Executed When a Packet Is Received
procedure SENDPACKET procedure SENDPACKET
if (RLWND > RLWNDPrevious) or (RLWND - RLWNDPrevious < ackedBytes) if (RLWND > RLWNDPrevious) or (RLWND - RLWNDPrevious < ackedBytes)
then then
RLWNDPrevious = RLWND RLWNDPrevious = RLWND
else else
RLWNDPrevious = RLWND - ackedBytes RLWNDPrevious = RLWND - ackedBytes
end if end if
ackedBytes = 0 ackedBytes = 0
RLWNDPrevious = RLWND RLWNDPrevious = RLWND
//Compute the RWND to include in the packet //Compute the RLWND to include in the packet
RLWND = min(RLWND, fcwnd) RLWND = min(RLWND, fcwnd)
end procedure end procedure
Figure 3: Procedure executed when a packet is sent Figure 3: Procedure Executed When a Packet Is Sent
Acknowledgments
This work was supported by the EU through the StandICT projects RXQ,
CCI, and CEL6; the NGI Pointer RIM project; and the H2020 5G-RANGE
project; and by the Spanish Ministry of Economy and Competitiveness
through the 5G-City project (TEC2016-76795-C6-3-R).
We would like to thank ICCRG chairs Reese Enghardt and Vidhi Goel for
their support on this work. We would also like to thank Daniel Havey
for his help. We would like to thank Colin Perkins, Mirja Kühlewind,
and Vidhi Goel for their reviews and comments on earlier draft
versions of this document.
Authors' Addresses Authors' Addresses
Marcelo Bagnulo Marcelo Bagnulo
Universidad Carlos III de Madrid Universidad Carlos III de Madrid
Email: marcelo@it.uc3m.es Email: marcelo@it.uc3m.es
Alberto Garcia-Martinez Alberto García-Martínez
Universidad Carlos III de Madrid Universidad Carlos III de Madrid
Email: alberto@it.uc3m.es Email: alberto@it.uc3m.es
Gabriel Montenegro Gabriel Montenegro
Email: g.e.montenegro@hotmail.com Email: g.e.montenegro@hotmail.com
Praveen Balasubramanian Praveen Balasubramanian
Confluent Confluent
Email: pravb.ietf@gmail.com Email: pravb.ietf@gmail.com
 End of changes. 154 change blocks. 
508 lines changed or deleted 563 lines changed or added

This html diff was produced by rfcdiff 1.48.