draft-ietf-rmcat-coupled-cc-07.txt   draft-ietf-rmcat-coupled-cc-08.txt 
RTP Media Congestion Avoidance Techniques (rmcat) S. Islam RTP Media Congestion Avoidance Techniques (rmcat) S. Islam
Internet-Draft M. Welzl Internet-Draft M. Welzl
Intended status: Experimental S. Gjessing Intended status: Experimental S. Gjessing
Expires: March 19, 2018 University of Oslo Expires: July 14, 2019 University of Oslo
September 15, 2017 January 10, 2019
Coupled congestion control for RTP media Coupled congestion control for RTP media
draft-ietf-rmcat-coupled-cc-07 draft-ietf-rmcat-coupled-cc-08
Abstract Abstract
When multiple congestion controlled Real-time Transport Protocol When multiple congestion controlled Real-time Transport Protocol
(RTP) sessions traverse the same network bottleneck, combining their (RTP) sessions traverse the same network bottleneck, combining their
controls can improve the total on-the-wire behavior in terms of controls can improve the total on-the-wire behavior in terms of
delay, loss and fairness. This document describes such a method for delay, loss and fairness. This document describes such a method for
flows that have the same sender, in a way that is as flexible and flows that have the same sender, in a way that is as flexible and
simple as possible while minimizing the amount of changes needed to simple as possible while minimizing the amount of changes needed to
existing RTP applications. It specifies how to apply the method for existing RTP applications. It specifies how to apply the method for
skipping to change at page 1, line 33 skipping to change at page 1, line 33
congestion control algorithms. congestion control algorithms.
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79. provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/. Drafts is at https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
This Internet-Draft will expire on March 19, 2018. This Internet-Draft will expire on July 14, 2019.
Copyright Notice Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the Copyright (c) 2019 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 Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of (https://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License. described in the Simplified BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . 4 3. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . 4
4. Architectural overview . . . . . . . . . . . . . . . . . . . 5 4. Architectural overview . . . . . . . . . . . . . . . . . . . 5
5. Roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5. Roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1. SBD . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5.1. SBD . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.2. FSE . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 5.2. FSE . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5.3. Flows . . . . . . . . . . . . . . . . . . . . . . . . . . 8 5.3. Flows . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5.3.1. Example algorithm 1 - Active FSE . . . . . . . . . . 9 5.3.1. Example algorithm 1 - Active FSE . . . . . . . . . . 9
5.3.2. Example algorithm 2 - Conservative Active FSE . . . . 10 5.3.2. Example algorithm 2 - Conservative Active FSE . . . . 11
6. Application . . . . . . . . . . . . . . . . . . . . . . . . . 11 6. Application . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.1. NADA . . . . . . . . . . . . . . . . . . . . . . . . . . 11 6.1. NADA . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6.2. General recommendations . . . . . . . . . . . . . . . . . 11 6.2. General recommendations . . . . . . . . . . . . . . . . . 12
7. Expected feedback from experiments . . . . . . . . . . . . . 12 7. Expected feedback from experiments . . . . . . . . . . . . . 12
8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 12 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 13
9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 13 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 13
10. Security Considerations . . . . . . . . . . . . . . . . . . . 13 10. Security Considerations . . . . . . . . . . . . . . . . . . . 13
11. References . . . . . . . . . . . . . . . . . . . . . . . . . 13 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 14
11.1. Normative References . . . . . . . . . . . . . . . . . . 13 11.1. Normative References . . . . . . . . . . . . . . . . . . 14
11.2. Informative References . . . . . . . . . . . . . . . . . 14 11.2. Informative References . . . . . . . . . . . . . . . . . 14
Appendix A. Application to GCC . . . . . . . . . . . . . . . . . 15 Appendix A. Application to GCC . . . . . . . . . . . . . . . . . 16
Appendix B. Scheduling . . . . . . . . . . . . . . . . . . . . . 16 Appendix B. Scheduling . . . . . . . . . . . . . . . . . . . . . 16
Appendix C. Example algorithm - Passive FSE . . . . . . . . . . 16 Appendix C. Example algorithm - Passive FSE . . . . . . . . . . 16
C.1. Example operation (passive) . . . . . . . . . . . . . . . 19 C.1. Example operation (passive) . . . . . . . . . . . . . . . 19
Appendix D. Change log . . . . . . . . . . . . . . . . . . . . . 23 Appendix D. Change log . . . . . . . . . . . . . . . . . . . . . 23
D.1. draft-welzl-rmcat-coupled-cc . . . . . . . . . . . . . . 23 D.1. draft-welzl-rmcat-coupled-cc . . . . . . . . . . . . . . 23
D.1.1. Changes from -00 to -01 . . . . . . . . . . . . . . . 23 D.1.1. Changes from -00 to -01 . . . . . . . . . . . . . . . 23
D.1.2. Changes from -01 to -02 . . . . . . . . . . . . . . . 23 D.1.2. Changes from -01 to -02 . . . . . . . . . . . . . . . 23
D.1.3. Changes from -02 to -03 . . . . . . . . . . . . . . . 23 D.1.3. Changes from -02 to -03 . . . . . . . . . . . . . . . 23
D.1.4. Changes from -03 to -04 . . . . . . . . . . . . . . . 24 D.1.4. Changes from -03 to -04 . . . . . . . . . . . . . . . 24
D.1.5. Changes from -04 to -05 . . . . . . . . . . . . . . . 24 D.1.5. Changes from -04 to -05 . . . . . . . . . . . . . . . 24
D.2. draft-ietf-rmcat-coupled-cc . . . . . . . . . . . . . . . 24 D.2. draft-ietf-rmcat-coupled-cc . . . . . . . . . . . . . . . 24
D.2.1. Changes from draft-welzl-rmcat-coupled-cc-05 . . . . 24 D.2.1. Changes from draft-welzl-rmcat-coupled-cc-05 . . . . 24
D.2.2. Changes from -00 to -01 . . . . . . . . . . . . . . . 24 D.2.2. Changes from -00 to -01 . . . . . . . . . . . . . . . 24
D.2.3. Changes from -01 to -02 . . . . . . . . . . . . . . . 24 D.2.3. Changes from -01 to -02 . . . . . . . . . . . . . . . 24
D.2.4. Changes from -02 to -03 . . . . . . . . . . . . . . . 24 D.2.4. Changes from -02 to -03 . . . . . . . . . . . . . . . 24
D.2.5. Changes from -03 to -04 . . . . . . . . . . . . . . . 24 D.2.5. Changes from -03 to -04 . . . . . . . . . . . . . . . 25
D.2.6. Changes from -04 to -05 . . . . . . . . . . . . . . . 25 D.2.6. Changes from -04 to -05 . . . . . . . . . . . . . . . 25
D.2.7. Changes from -05 to -06 . . . . . . . . . . . . . . . 25 D.2.7. Changes from -05 to -06 . . . . . . . . . . . . . . . 25
D.2.8. Changes from -06 to -07 . . . . . . . . . . . . . . . 25 D.2.8. Changes from -06 to -07 . . . . . . . . . . . . . . . 25
D.2.9. Changes from -07 to -08 . . . . . . . . . . . . . . . 25
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 25 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 25
1. Introduction 1. Introduction
When there is enough data to send, a congestion controller attempts When there is enough data to send, a congestion controller attempts
to increase its sending rate until the path's capacity has been to increase its sending rate until the path's capacity has been
reached. Some controllers detect path capacity by increasing the reached. Some controllers detect path capacity by increasing the
sending rate further, until packets are ECN-marked [RFC8087] or sending rate further, until packets are ECN-marked [RFC8087] or
dropped, and then decreasing the sending rate until that stops dropped, and then decreasing the sending rate until that stops
happening. This process inevitably creates undesirable queuing delay happening. This process inevitably creates undesirable queuing delay
when multiple congestion-controlled connections traverse the same when multiple congestion-controlled connections traverse the same
network bottleneck, and each connection overshoots the path capacity network bottleneck, and each connection overshoots the path capacity
as it determines its sending rate. as it determines its sending rate.
The Congestion Manager (CM) [RFC3124] couples flows by providing a The Congestion Manager (CM) [RFC3124] couples flows by providing a
single congestion controller. It is hard to implement because it single congestion controller. It is hard to implement because it
requires an additional congestion controller and removes all per- requires an additional congestion controller and removes all per-
connection congestion control functionality, which is quite a connection congestion control functionality, which is quite a
skipping to change at page 4, line 30 skipping to change at page 4, line 31
Shared Bottleneck Detection (SBD): Shared Bottleneck Detection (SBD):
The entity that determines which flows traverse the same The entity that determines which flows traverse the same
bottleneck in the network, or the process of doing so. bottleneck in the network, or the process of doing so.
3. Limitations 3. Limitations
Sender-side only: Sender-side only:
Shared bottlenecks can exist when multiple flows originate from Shared bottlenecks can exist when multiple flows originate from
the same sender, or when flows from different senders reach the the same sender, or when flows from different senders reach the
same receiver (see [I-D.ietf-rmcat-sbd], section 3). Coupled same receiver (see [RFC8382], section 3). Coupled congestion
congestion control as described here only supports the former control as described here only supports the former case, not
case, not the latter, as it operates inside a single host on the latter, as it operates inside a single host on the sender
the sender side. side.
Shared bottlenecks do not change quickly: Shared bottlenecks do not change quickly:
As per the definition above, a bottleneck depends on cross As per the definition above, a bottleneck depends on cross
traffic, and since such traffic can heavily fluctuate, traffic, and since such traffic can heavily fluctuate,
bottlenecks can change at a high frequency (e.g., there can be bottlenecks can change at a high frequency (e.g., there can be
oscillation between two or more links). This means that, when oscillation between two or more links). This means that, when
flows are partially routed along different paths, they may flows are partially routed along different paths, they may
quickly change between sharing and not sharing a bottleneck. quickly change between sharing and not sharing a bottleneck.
For simplicity, here it is assumed that a shared bottleneck is For simplicity, here it is assumed that a shared bottleneck is
valid for a time interval that is significantly longer than the valid for a time interval that is significantly longer than the
skipping to change at page 7, line 25 skipping to change at page 7, line 25
2. Via configuration: e.g. by assuming that a common wireless uplink 2. Via configuration: e.g. by assuming that a common wireless uplink
is also a shared bottleneck. is also a shared bottleneck.
3. From measurements: e.g. by considering correlations among 3. From measurements: e.g. by considering correlations among
measured delay and loss as an indication of a shared bottleneck. measured delay and loss as an indication of a shared bottleneck.
The methods above have some essential trade-offs: e.g., multiplexing The methods above have some essential trade-offs: e.g., multiplexing
is a completely reliable measure, however it is limited in scope to is a completely reliable measure, however it is limited in scope to
two end points (i.e., it cannot be applied to couple congestion two end points (i.e., it cannot be applied to couple congestion
controllers of one sender talking to multiple receivers). A controllers of one sender talking to multiple receivers). A
measurement-based SBD mechanism is described in [I-D.ietf-rmcat-sbd]. measurement-based SBD mechanism is described in [RFC8382].
Measurements can never be 100% reliable, in particular because they Measurements can never be 100% reliable, in particular because they
are based on the past but applying coupled congestion control means are based on the past but applying coupled congestion control means
to make an assumption about the future; it is therefore recommended to make an assumption about the future; it is therefore recommended
to implement cautionary measures, e.g. by disabling coupled to implement cautionary measures, e.g. by disabling coupled
congestion control if enabling it causes a significant increase in congestion control if enabling it causes a significant increase in
delay and/or packet loss. Measurements also take time, which entails delay and/or packet loss. Measurements also take time, which entails
a certain delay for turning on coupling (refer to a certain delay for turning on coupling (refer to [RFC8382] for
[I-D.ietf-rmcat-sbd] for details). Using system configuration to details). Using system configuration to decide about shared
decide about shared bottlenecks can be more efficient (faster to bottlenecks can be more efficient (faster to obtain) than using
obtain) than using measurements, but it relies on assumptions about measurements, but it relies on assumptions about the network
the network environment. environment.
5.2. FSE 5.2. FSE
The FSE contains a list of all flows that have registered with it. The FSE contains a list of all flows that have registered with it.
For each flow, it stores the following: For each flow, it stores the following:
o a unique flow number f to identify the flow. o a unique flow number f to identify the flow.
o the FGI of the FG that it belongs to (based on the definitions in o the FGI of the FG that it belongs to (based on the definitions in
this document, a flow has only one bottleneck, and can therefore this document, a flow has only one bottleneck, and can therefore
be in only one FG). be in only one FG).
o a priority P(f), which is a positive number, greater than zero. o a priority P(f), which is a positive number, greater than zero.
o The rate used by the flow in bits per second, FSE_R(f). o The rate used by the flow in bits per second, FSE_R(f).
o The desired rate DR(f) of flow f. This can be smaller than
FSE_R(f) if the application feeding into the flow has less data to
send than FSE_R(f) would allow, or if a maximum value is imposed
on the rate. In the absence of such limits DR(f) must be set to
the sending rate provided by the congestion control module of flow
f.
Note that the absolute range of priorities does not matter: the Note that the absolute range of priorities does not matter: the
algorithm works with a flow's priority portion of the sum of all algorithm works with a flow's priority portion of the sum of all
priority values. For example, if there are two flows, flow 1 with priority values. For example, if there are two flows, flow 1 with
priority 1 and flow 2 with priority 2, the sum of the priorities is priority 1 and flow 2 with priority 2, the sum of the priorities is
3. Then, flow 1 will be assigned 1/3 of the aggregate sending rate 3. Then, flow 1 will be assigned 1/3 of the aggregate sending rate
and flow 2 will be assigned 2/3 of the aggregate sending rate. and flow 2 will be assigned 2/3 of the aggregate sending rate.
Priorities can be mapped to the "very-low", "low", "medium" or "high" Priorities can be mapped to the "very-low", "low", "medium" or "high"
priority levels described in [I-D.ietf-rtcweb-transports] by simply priority levels described in [I-D.ietf-rtcweb-transports] by simply
using the values 1, 2, 4 and 8, respectively. using the values 1, 2, 4 and 8, respectively.
skipping to change at page 8, line 44 skipping to change at page 8, line 51
Below, two example algorithms are described. While other algorithms Below, two example algorithms are described. While other algorithms
could be used instead, the same algorithm must be applied to all could be used instead, the same algorithm must be applied to all
flows. Names of variables used in the algorithms are explained flows. Names of variables used in the algorithms are explained
below. below.
o CC_R(f) - The rate received from the congestion controller of flow o CC_R(f) - The rate received from the congestion controller of flow
f when it calls UPDATE. f when it calls UPDATE.
o FSE_R(f) - The rate calculated by the FSE for flow f. o FSE_R(f) - The rate calculated by the FSE for flow f.
o DR(f) - The desired rate of flow f.
o S_CR - The sum of the calculated rates of all flows in the same o S_CR - The sum of the calculated rates of all flows in the same
FG; this value is used to calculate the sending rate. FG; this value is used to calculate the sending rate.
o FG - A group of flows having the same FGI, and hence sharing the o FG - A group of flows having the same FGI, and hence sharing the
same bottleneck. same bottleneck.
o P(f) - The priority of flow f which is received from the flow's o P(f) - The priority of flow f which is received from the flow's
congestion controller; the FSE uses this variable for calculating congestion controller; the FSE uses this variable for calculating
FSE_R(f). FSE_R(f).
o S_P - The sum of all the priorities. o S_P - The sum of all the priorities.
o TLO - The total leftover rate: the sum of rates that could not be
assigned to flows that were limited by their desired rate.
o S_P2 - The sum of all the priorities of flows to which a share of
the TLO can be assigned.
5.3.1. Example algorithm 1 - Active FSE 5.3.1. Example algorithm 1 - Active FSE
This algorithm was designed to be the simplest possible method to This algorithm was designed to be the simplest possible method to
assign rates according to the priorities of flows. Simulations assign rates according to the priorities of flows. Simulations
results in [fse] indicate that it does however not significantly results in [fse] indicate that it does however not significantly
reduce queuing delay and packet loss. reduce queuing delay and packet loss.
(1) When a flow f starts, it registers itself with SBD and the FSE. (1) When a flow f starts, it registers itself with SBD and the FSE.
FSE_R(f) is initialized with the congestion controller's initial FSE_R(f) is initialized with the congestion controller's initial
rate. SBD will assign the correct FGI. When a flow is assigned rate. SBD will assign the correct FGI. When a flow is assigned
an FGI, it adds its FSE_R(f) to S_CR. an FGI, it adds its FSE_R(f) to S_CR.
(2) When a flow f stops or pauses, its entry is removed from the (2) When a flow f stops or pauses, its entry is removed from the
list. list.
(3) Every time the congestion controller of the flow f determines a (3) Every time the congestion controller of the flow f determines a
new sending rate CC_R(f), the flow calls UPDATE, which carries new sending rate CC_R(f), the flow calls UPDATE, which carries
out the tasks listed below to derive the new sending rates for out the tasks listed below to derive the new sending rates for
all the flows in the FG. A flow's UPDATE function uses a local all the flows in the FG. A flow's UPDATE function uses three
(i.e. per-flow) temporary variable S_P, which is the sum of all local (i.e. per-flow) temporary variables: S_P, TLO and S_P2.
the priorities.
(a) It updates S_CR. (a) It updates S_CR.
S_CR = S_CR + CC_R(f) - FSE_R(f) S_CR = S_CR + CC_R(f) - FSE_R(f)
(b) It calculates the sum of all the priorities, S_P. (b) It calculates the sum of all the priorities, S_P.
S_P = 0 S_P = 0
for all flows i in FG do for all flows i in FG do
S_P = S_P + P(i) S_P = S_P + P(i)
end for end for
(c) It calculates the sending rates for all the flows in an FG (c) It calculates the sending rates for all the flows in an FG,
and distributes them. the total leftover rate (TLO) from flows that are limited
by their desired rate, and the sum of the priorities of all
other flows, S_P2.
TLO = 0
S_P2 = 0
for all flows i in FG do for all flows i in FG do
FSE_R(i) = (P(i)*S_CR)/S_P FSE_R(i) = P(i) * S_CR /S_P
if FSE_R(i) >= DR(i) then
TLO = TLO + FSE_R(i) - DR(i)
FSE_R(i) = DR(i)
else
S_P2 = S_P2 + P(i)
end if
end for
(d) It checks if there are flows that are limited by their DR
and cannot accept their share of the TLO, and updates TLO
and S_P2 accordingly.
for all flows i in FG do
if FSE_R(i) < DR(i) then
if FSE_R(i) + TLO * P(i) / S_P2 > DR(i) then
TLO = TLO - ( DR(i) - FSE_R(i) )
FSE_R(i) = DR(i)
S_P2 = S_P2 - P(i)
end if
end if
end for
(e) It assigns the non-limited flow their share of the total
leftover rate and sends all the rates to all the flows.
for all flows i in FG do
if FSE_R(i) < DR(i) then
FSE_R(i) = FSE_R(i) + P(i) * TLO / S_P2
end if
send FSE_R(i) to the flow i send FSE_R(i) to the flow i
end for end for
5.3.2. Example algorithm 2 - Conservative Active FSE 5.3.2. Example algorithm 2 - Conservative Active FSE
This algorithm extends algorithm 1 to conservatively emulate the This algorithm changes algorithm 1 to conservatively emulate the
behavior of a single flow by proportionally reducing the aggregate behavior of a single flow by proportionally reducing the aggregate
rate on congestion. Simulations results in [fse] indicate that it rate on congestion. Simulations results in [fse] indicate that it
can significantly reduce queuing delay and packet loss. can significantly reduce queuing delay and packet loss.
(1) When a flow f starts, it registers itself with SBD and the FSE. Step (a) of the UPDATE function is changed as described below. This
FSE_R(f) is initialized with the congestion controller's initial also introduces a local variable DELTA, which is used to calculate
rate. SBD will assign the correct FGI. When a flow is assigned the difference between CC_R(f) and the previously stored FSE_R(f).
an FGI, it adds its FSE_R(f) to S_CR. To prevent flows from either ignoring congestion or overreacting, a
timer keeps them from changing their rates immediately after the
(2) When a flow f stops or pauses, its entry is removed from the common rate reduction that follows a congestion event. This timer is
list. set to 2 RTTs of the flow that experienced congestion because it is
assumed that a congestion event can persist for up to one RTT of that
(3) Every time the congestion controller of the flow f determines a flow, with another RTT added to compensate for fluctuations in the
new sending rate CC_R(f), the flow calls UPDATE, which carries measured RTT value.
out the tasks listed below to derive the new sending rates for
all the flows in the FG. A flow's UPDATE function uses a local
(i.e. per-flow) temporary variable S_P, which is the sum of all
the priorities, and a local variable DELTA, which is used to
calculate the difference between CC_R(f) and the previously
stored FSE_R(f). To prevent flows from either ignoring
congestion or overreacting, a timer keeps them from changing
their rates immediately after the common rate reduction that
follows a congestion event. This timer is set to 2 RTTs of the
flow that experienced congestion because it is assumed that a
congestion event can persist for up to one RTT of that flow,
with another RTT added to compensate for fluctuations in the
measured RTT value.
(a) It updates S_CR based on DELTA. (a) It updates S_CR based on DELTA.
if Timer has expired or not set then if Timer has expired or was not set then
DELTA = CC_R(f) - FSE_R(f) DELTA = CC_R(f) - FSE_R(f)
if DELTA < 0 then // Reduce S_CR proportionally if DELTA < 0 then // Reduce S_CR proportionally
S_CR = S_CR * CC_R(f) / FSE_R(f) S_CR = S_CR * CC_R(f) / FSE_R(f)
Set Timer for 2 RTTs Set Timer for 2 RTTs
else else
S_CR = S_CR + DELTA S_CR = S_CR + DELTA
end if end if
end if end if
(b) It calculates the sum of all the priorities, S_P.
S_P = 0
for all flows i in FG do
S_P = S_P + P(i)
end for
(c) It calculates the sending rates for all the flows in an FG
and distributes them.
for all flows i in FG do
FSE_R(i) = (P(i)*S_CR)/S_P
send FSE_R(i) to the flow i
end for
6. Application 6. Application
This section specifies how the FSE can be applied to specific This section specifies how the FSE can be applied to specific
congestion control mechanisms and makes general recommendations that congestion control mechanisms and makes general recommendations that
facilitate applying the FSE to future congestion controls. facilitate applying the FSE to future congestion controls.
6.1. NADA 6.1. NADA
Network-Assisted Dynamic Adapation (NADA) [I-D.ietf-rmcat-nada] is a Network-Assisted Dynamic Adapation (NADA) [I-D.ietf-rmcat-nada] is a
congestion control scheme for rtcweb. It calculates a reference rate congestion control scheme for rtcweb. It calculates a reference rate
skipping to change at page 12, line 44 skipping to change at page 13, line 13
testers are invited to document their findings in an Internet draft. testers are invited to document their findings in an Internet draft.
8. Acknowledgements 8. Acknowledgements
This document has benefitted from discussions with and feedback from This document has benefitted from discussions with and feedback from
Andreas Petlund, Anna Brunstrom, Colin Perkins, David Hayes, David Andreas Petlund, Anna Brunstrom, Colin Perkins, David Hayes, David
Ros (who also gave the FSE its name), Ingemar Johansson, Karen Ros (who also gave the FSE its name), Ingemar Johansson, Karen
Nielsen, Kristian Hiorth, Mirja Kuehlewind, Martin Stiemerling, Nielsen, Kristian Hiorth, Mirja Kuehlewind, Martin Stiemerling,
Spencer Dawkins, Varun Singh, Xiaoqing Zhu, and Zaheduzzaman Sarker. Spencer Dawkins, Varun Singh, Xiaoqing Zhu, and Zaheduzzaman Sarker.
The authors would like to especially thank Xiaoqing Zhu and Stefan The authors would like to especially thank Xiaoqing Zhu and Stefan
Holmer for helping with NADA and GCC. Holmer for helping with NADA and GCC, and Julius Flohr for helping us
correct the active algorithm for the case of application-limited
flows.
This work was partially funded by the European Community under its This work was partially funded by the European Community under its
Seventh Framework Programme through the Reducing Internet Transport Seventh Framework Programme through the Reducing Internet Transport
Latency (RITE) project (ICT-317700). Latency (RITE) project (ICT-317700).
9. IANA Considerations 9. IANA Considerations
This memo includes no request to IANA. This memo includes no request to IANA.
10. Security Considerations 10. Security Considerations
skipping to change at page 13, line 40 skipping to change at page 14, line 10
eliminated, and no major harm is done. eliminated, and no major harm is done.
Implementers should also be aware of the Security Considerations Implementers should also be aware of the Security Considerations
sections of [RFC3124], [RFC5348], and [RFC7478]. sections of [RFC3124], [RFC5348], and [RFC7478].
11. References 11. References
11.1. Normative References 11.1. Normative References
[I-D.ietf-rmcat-nada] [I-D.ietf-rmcat-nada]
Zhu, X., Pan, R., Ramalho, M., Cruz, S., Jones, P., Fu, Zhu, X., *, R., Ramalho, M., Cruz, S., Jones, P., Fu, J.,
J., and S. D'Aronco, "NADA: A Unified Congestion Control and S. D'Aronco, "NADA: A Unified Congestion Control
Scheme for Real-Time Media", draft-ietf-rmcat-nada-04 Scheme for Real-Time Media", draft-ietf-rmcat-nada-09
(work in progress), March 2017. (work in progress), August 2018.
[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, DOI 10.17487/ Requirement Levels", BCP 14, RFC 2119,
RFC2119, March 1997, <https://www.rfc-editor.org/info/ DOI 10.17487/RFC2119, March 1997,
rfc2119>. <https://www.rfc-editor.org/info/rfc2119>.
[RFC3124] Balakrishnan, H. and S. Seshan, "The Congestion Manager", [RFC3124] Balakrishnan, H. and S. Seshan, "The Congestion Manager",
RFC 3124, DOI 10.17487/RFC3124, June 2001, RFC 3124, DOI 10.17487/RFC3124, June 2001,
<https://www.rfc-editor.org/info/rfc3124>. <https://www.rfc-editor.org/info/rfc3124>.
[RFC5348] Floyd, S., Handley, M., Padhye, J., and J. Widmer, "TCP [RFC5348] Floyd, S., Handley, M., Padhye, J., and J. Widmer, "TCP
Friendly Rate Control (TFRC): Protocol Specification", RFC Friendly Rate Control (TFRC): Protocol Specification",
5348, DOI 10.17487/RFC5348, September 2008, RFC 5348, DOI 10.17487/RFC5348, September 2008,
<https://www.rfc-editor.org/info/rfc5348>. <https://www.rfc-editor.org/info/rfc5348>.
11.2. Informative References 11.2. Informative References
[anrw2016] [anrw2016]
Islam, S. and M. Welzl, "Start Me Up:Determining and Islam, S. and M. Welzl, "Start Me Up:Determining and
Sharing TCP's Initial Congestion Window", ACM, IRTF, ISOC Sharing TCP's Initial Congestion Window", ACM, IRTF, ISOC
Applied Networking Research Workshop 2016 (ANRW 2016) , Applied Networking Research Workshop 2016 (ANRW 2016) ,
2016. 2016.
skipping to change at page 14, line 34 skipping to change at page 15, line 8
2014. 2014.
[fse-noms] [fse-noms]
Islam, S., Welzl, M., Hayes, D., and S. Gjessing, Islam, S., Welzl, M., Hayes, D., and S. Gjessing,
"Managing Real-Time Media Flows through a Flow State "Managing Real-Time Media Flows through a Flow State
Exchange", IEEE NOMS 2016, Istanbul, Turkey , 2016. Exchange", IEEE NOMS 2016, Istanbul, Turkey , 2016.
[I-D.ietf-rmcat-eval-test] [I-D.ietf-rmcat-eval-test]
Sarker, Z., Singh, V., Zhu, X., and M. Ramalho, "Test Sarker, Z., Singh, V., Zhu, X., and M. Ramalho, "Test
Cases for Evaluating RMCAT Proposals", draft-ietf-rmcat- Cases for Evaluating RMCAT Proposals", draft-ietf-rmcat-
eval-test-05 (work in progress), April 2017. eval-test-08 (work in progress), November 2018.
[I-D.ietf-rmcat-gcc] [I-D.ietf-rmcat-gcc]
Holmer, S., Lundin, H., Carlucci, G., Cicco, L., and S. Holmer, S., Lundin, H., Carlucci, G., Cicco, L., and S.
Mascolo, "A Google Congestion Control Algorithm for Real- Mascolo, "A Google Congestion Control Algorithm for Real-
Time Communication", draft-ietf-rmcat-gcc-02 (work in Time Communication", draft-ietf-rmcat-gcc-02 (work in
progress), July 2016. progress), July 2016.
[I-D.ietf-rmcat-sbd]
Hayes, D., Ferlin, S., Welzl, M., and K. Hiorth, "Shared
Bottleneck Detection for Coupled Congestion Control for
RTP Media.", draft-ietf-rmcat-sbd-08 (work in progress),
July 2017.
[I-D.ietf-rtcweb-overview] [I-D.ietf-rtcweb-overview]
Alvestrand, H., "Overview: Real Time Protocols for Alvestrand, H., "Overview: Real Time Protocols for
Browser-based Applications", draft-ietf-rtcweb-overview-18 Browser-based Applications", draft-ietf-rtcweb-overview-19
(work in progress), March 2017. (work in progress), November 2017.
[I-D.ietf-rtcweb-transports] [I-D.ietf-rtcweb-transports]
Alvestrand, H., "Transports for WebRTC", Internet-draft Alvestrand, H., "Transports for WebRTC", Internet-draft
draft-ietf-rtcweb-transports-17.txt, October 2016. draft-ietf-rtcweb-transports-17.txt, October 2016.
[IETF-93] Islam, S., Welzl, M., and S. Gjessing, "Updates on Coupled [IETF-93] Islam, S., Welzl, M., and S. Gjessing, "Updates on Coupled
Congestion Control for RTP Media", July 2015, Congestion Control for RTP Media", July 2015,
<https://www.ietf.org/proceedings/93/rmcat.html>. <https://www.ietf.org/proceedings/93/rmcat.html>.
[IETF-94] Islam, S., Welzl, M., and S. Gjessing, "Updates on Coupled [IETF-94] Islam, S., Welzl, M., and S. Gjessing, "Updates on Coupled
Congestion Control for RTP Media", November 2015, Congestion Control for RTP Media", November 2015,
<https://www.ietf.org/proceedings/94/rmcat.html>. <https://www.ietf.org/proceedings/94/rmcat.html>.
[RFC7478] Holmberg, C., Hakansson, S., and G. Eriksson, "Web Real- [RFC7478] Holmberg, C., Hakansson, S., and G. Eriksson, "Web Real-
Time Communication Use Cases and Requirements", RFC 7478, Time Communication Use Cases and Requirements", RFC 7478,
DOI 10.17487/RFC7478, March 2015, <https://www.rfc- DOI 10.17487/RFC7478, March 2015,
editor.org/info/rfc7478>. <https://www.rfc-editor.org/info/rfc7478>.
[RFC7656] Lennox, J., Gross, K., Nandakumar, S., Salgueiro, G., and [RFC7656] Lennox, J., Gross, K., Nandakumar, S., Salgueiro, G., and
B. Burman, Ed., "A Taxonomy of Semantics and Mechanisms B. Burman, Ed., "A Taxonomy of Semantics and Mechanisms
for Real-Time Transport Protocol (RTP) Sources", RFC 7656, for Real-Time Transport Protocol (RTP) Sources", RFC 7656,
DOI 10.17487/RFC7656, November 2015, <https://www.rfc- DOI 10.17487/RFC7656, November 2015,
editor.org/info/rfc7656>. <https://www.rfc-editor.org/info/rfc7656>.
[RFC8087] Fairhurst, G. and M. Welzl, "The Benefits of Using [RFC8087] Fairhurst, G. and M. Welzl, "The Benefits of Using
Explicit Congestion Notification (ECN)", RFC 8087, DOI Explicit Congestion Notification (ECN)", RFC 8087,
10.17487/RFC8087, March 2017, <https://www.rfc- DOI 10.17487/RFC8087, March 2017,
editor.org/info/rfc8087>. <https://www.rfc-editor.org/info/rfc8087>.
[RFC8382] Hayes, D., Ed., Ferlin, S., Welzl, M., and K. Hiorth,
"Shared Bottleneck Detection for Coupled Congestion
Control for RTP Media", RFC 8382, DOI 10.17487/RFC8382,
June 2018, <https://www.rfc-editor.org/info/rfc8382>.
[rtcweb-rtp-usage] [rtcweb-rtp-usage]
Perkins, C., Westerlund, M., and J. Ott, "Web Real-Time Perkins, C., Westerlund, M., and J. Ott, "Web Real-Time
Communication (WebRTC): Media Transport and Use of RTP", Communication (WebRTC): Media Transport and Use of RTP",
Internet-draft draft-ietf-rtcweb-rtp-usage-26.txt, March Internet-draft draft-ietf-rtcweb-rtp-usage-26.txt, March
2016. 2016.
[transport-multiplex] [transport-multiplex]
Westerlund, M. and C. Perkins, "Multiple RTP Sessions on a Westerlund, M. and C. Perkins, "Multiple RTP Sessions on a
Single Lower-Layer Transport", Internet-draft draft- Single Lower-Layer Transport", Internet-draft draft-
skipping to change at page 16, line 40 skipping to change at page 17, line 12
update and react to the overall FSE state more often than longer-RTT update and react to the overall FSE state more often than longer-RTT
flows, which can produce unwanted side effects. This problem is more flows, which can produce unwanted side effects. This problem is more
significant when the congestion control convergence depends on the significant when the congestion control convergence depends on the
RTT. While the passive algorithm works better for congestion RTT. While the passive algorithm works better for congestion
controls with RTT-independent convergence, it can still produce controls with RTT-independent convergence, it can still produce
oscillations on short time scales. The algorithm described below is oscillations on short time scales. The algorithm described below is
therefore considered as highly experimental and not safe to deploy therefore considered as highly experimental and not safe to deploy
outside of testbed environments. Results of a simplified passive FSE outside of testbed environments. Results of a simplified passive FSE
algorithm with both NADA and GCC can be found in [fse-noms]. algorithm with both NADA and GCC can be found in [fse-noms].
This passive version of the FSE stores the following information in In the passive version of the FSE, TLO (the Total Leftover Rate) is a
addition to the variables described in Section 5.2: static variable per FG which is initialized to 0. Additionally, S_CR
is limited to increase or decrease as conservatively as a flow's
o The desired rate DR(f) of flow f. This can be smaller than the congestion controller decides in order to prohibit sudden rate jumps.
calculated rate if the application feeding into the flow has less
data to send than the congestion controller would allow. In case
of a bulk transfer, DR(f) must be set to CC_R(f) received from the
congestion module of flow f.
The passive version of the FSE contains one static variable per FG
called TLO (Total Leftover Rate -- used to let a flow 'take'
bandwidth from application-limited or terminated flows) which is
initialized to 0. For the passive version, S_CR is limited to
increase or decrease as conservatively as a flow's congestion
controller decides in order to prohibit sudden rate jumps.
(1) When a flow f starts, it registers itself with SBD and the FSE. (1) When a flow f starts, it registers itself with SBD and the FSE.
FSE_R(f) and DR(f) are initialized with the congestion FSE_R(f) and DR(f) are initialized with the congestion
controller's initial rate. SBD will assign the correct FGI. controller's initial rate. SBD will assign the correct FGI.
When a flow is assigned an FGI, it adds its FSE_R(f) to S_CR. When a flow is assigned an FGI, it adds its FSE_R(f) to S_CR.
(2) When a flow f stops or pauses, it sets its DR(f) to 0 and sets (2) When a flow f stops or pauses, it sets its DR(f) to 0 and sets
P(f) to -1. P(f) to -1.
(3) Every time the congestion controller of the flow f determines a (3) Every time the congestion controller of the flow f determines a
skipping to change at page 25, line 29 skipping to change at page 25, line 36
coupling algorithm. coupling algorithm.
D.2.7. Changes from -05 to -06 D.2.7. Changes from -05 to -06
o Incorporated comments by Colin Perkins. o Incorporated comments by Colin Perkins.
D.2.8. Changes from -06 to -07 D.2.8. Changes from -06 to -07
o Addressed OPSDIR, SECDIR, GENART, AD and IESG comments. o Addressed OPSDIR, SECDIR, GENART, AD and IESG comments.
D.2.9. Changes from -07 to -08
o Updated the algorithms in section 5 to support application-limited
flows. Moved definition of Desired Rate from appendix to section
5. Updated references.
Authors' Addresses Authors' Addresses
Safiqul Islam Safiqul Islam
University of Oslo University of Oslo
PO Box 1080 Blindern PO Box 1080 Blindern
Oslo N-0316 Oslo N-0316
Norway Norway
Phone: +47 22 84 08 37 Phone: +47 22 84 08 37
Email: safiquli@ifi.uio.no Email: safiquli@ifi.uio.no
 End of changes. 41 change blocks. 
113 lines changed or deleted 129 lines changed or added

This html diff was produced by rfcdiff 1.47. The latest version is available from http://tools.ietf.org/tools/rfcdiff/