draft-ietf-rmcat-sbd-06.txt   draft-ietf-rmcat-sbd-07.txt 
RTP Media Congestion Avoidance Techniques D. Hayes, Ed. RTP Media Congestion Avoidance Techniques D. Hayes, Ed.
Internet-Draft S. Ferlin Internet-Draft S. Ferlin
Intended status: Experimental Simula Research Laboratory Intended status: Experimental Simula Research Laboratory
Expires: August 19, 2017 M. Welzl Expires: December 10, 2017 M. Welzl
K. Hiorth K. Hiorth
University of Oslo University of Oslo
February 15, 2017 June 8, 2017
Shared Bottleneck Detection for Coupled Congestion Control for RTP Shared Bottleneck Detection for Coupled Congestion Control for RTP
Media. Media.
draft-ietf-rmcat-sbd-06 draft-ietf-rmcat-sbd-07
Abstract Abstract
This document describes a mechanism to detect whether end-to-end data This document describes a mechanism to detect whether end-to-end data
flows share a common bottleneck. It relies on summary statistics flows share a common bottleneck. It relies on summary statistics
that are calculated based on continuous measurements and used as that are calculated based on continuous measurements and used as
input to a grouping algorithm that runs wherever the knowledge is input to a grouping algorithm that runs wherever the knowledge is
needed. This mechanism complements the coupled congestion control needed. This mechanism complements the coupled congestion control
mechanism in draft-ietf-rmcat-coupled-cc. mechanism in draft-ietf-rmcat-coupled-cc.
skipping to change at page 1, line 39 skipping to change at page 1, line 39
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 http://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 August 19, 2017. This Internet-Draft will expire on December 10, 2017.
Copyright Notice Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the Copyright (c) 2017 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 (http://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
skipping to change at page 2, line 27 skipping to change at page 2, line 27
1.2.3. Path lag . . . . . . . . . . . . . . . . . . . . . . 4 1.2.3. Path lag . . . . . . . . . . . . . . . . . . . . . . 4
2. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 4 2. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1. Parameters and their effect . . . . . . . . . . . . . . . 7 2.1. Parameters and their effect . . . . . . . . . . . . . . . 7
2.2. Recommended parameter values . . . . . . . . . . . . . . 8 2.2. Recommended parameter values . . . . . . . . . . . . . . 8
3. Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3. Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.1. SBD feedback requirements . . . . . . . . . . . . . . . . 9 3.1. SBD feedback requirements . . . . . . . . . . . . . . . . 9
3.1.1. Feedback when all the logic is placed at the sender . 9 3.1.1. Feedback when all the logic is placed at the sender . 9
3.1.2. Feedback when the statistics are calculated at the 3.1.2. Feedback when the statistics are calculated at the
receiver and SBD performed at the sender . . . . . . 10 receiver and SBD performed at the sender . . . . . . 10
3.1.3. Feedback when bottlenecks can be determined at both 3.1.3. Feedback when bottlenecks can be determined at both
senders and receivers . . . . . . . . . . . . . . . . 10 senders and receivers . . . . . . . . . . . . . . . . 11
3.2. Key metrics and their calculation . . . . . . . . . . . . 11 3.2. Key metrics and their calculation . . . . . . . . . . . . 11
3.2.1. Mean delay . . . . . . . . . . . . . . . . . . . . . 11 3.2.1. Mean delay . . . . . . . . . . . . . . . . . . . . . 11
3.2.2. Skewness estimate . . . . . . . . . . . . . . . . . . 11 3.2.2. Skewness estimate . . . . . . . . . . . . . . . . . . 11
3.2.3. Variability estimate . . . . . . . . . . . . . . . . 12 3.2.3. Variability estimate . . . . . . . . . . . . . . . . 12
3.2.4. Oscillation estimate . . . . . . . . . . . . . . . . 12 3.2.4. Oscillation estimate . . . . . . . . . . . . . . . . 12
3.2.5. Packet loss . . . . . . . . . . . . . . . . . . . . . 13 3.2.5. Packet loss . . . . . . . . . . . . . . . . . . . . . 13
3.3. Flow Grouping . . . . . . . . . . . . . . . . . . . . . . 13 3.3. Flow Grouping . . . . . . . . . . . . . . . . . . . . . . 13
3.3.1. Flow grouping algorithm . . . . . . . . . . . . . . . 13 3.3.1. Flow grouping algorithm . . . . . . . . . . . . . . . 13
3.3.2. Using the flow group signal . . . . . . . . . . . . . 15 3.3.2. Using the flow group signal . . . . . . . . . . . . . 16
4. Enhancements to the basic SBD algorithm . . . . . . . . . . . 15 4. Enhancements to the basic SBD algorithm . . . . . . . . . . . 17
4.1. Reducing lag and improving responsiveness . . . . . . . . 15 4.1. Reducing lag and improving responsiveness . . . . . . . . 17
4.1.1. Improving the response of the skewness estimate . . . 16 4.1.1. Improving the response of the skewness estimate . . . 18
4.1.2. Improving the response of the variability estimate . 18 4.1.2. Improving the response of the variability estimate . 20
4.2. Removing oscillation noise . . . . . . . . . . . . . . . 18 4.2. Removing oscillation noise . . . . . . . . . . . . . . . 20
5. Measuring OWD . . . . . . . . . . . . . . . . . . . . . . . . 19 5. Measuring OWD . . . . . . . . . . . . . . . . . . . . . . . . 21
5.1. Time stamp resolution . . . . . . . . . . . . . . . . . . 19 5.1. Time stamp resolution . . . . . . . . . . . . . . . . . . 21
5.2. Clock skew . . . . . . . . . . . . . . . . . . . . . . . 19 5.2. Clock skew . . . . . . . . . . . . . . . . . . . . . . . 21
6. Expected feedback from experiments . . . . . . . . . . . . . 19 6. Expected feedback from experiments . . . . . . . . . . . . . 21
7. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 20 7. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 22
8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 20 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 22
9. Security Considerations . . . . . . . . . . . . . . . . . . . 20 9. Security Considerations . . . . . . . . . . . . . . . . . . . 22
10. Change history . . . . . . . . . . . . . . . . . . . . . . . 20 10. Change history . . . . . . . . . . . . . . . . . . . . . . . 22
11. References . . . . . . . . . . . . . . . . . . . . . . . . . 21 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 23
11.1. Normative References . . . . . . . . . . . . . . . . . . 21 11.1. Normative References . . . . . . . . . . . . . . . . . . 23
11.2. Informative References . . . . . . . . . . . . . . . . . 21 11.2. Informative References . . . . . . . . . . . . . . . . . 23
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 22 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 25
1. Introduction 1. Introduction
In the Internet, it is not normally known if flows (e.g., TCP In the Internet, it is not normally known if flows (e.g., TCP
connections or UDP data streams) traverse the same bottlenecks. Even connections or UDP data streams) traverse the same bottlenecks. Even
flows that have the same sender and receiver may take different paths flows that have the same sender and receiver may take different paths
and may or may not share a bottleneck. Flows that share a bottleneck and may or may not share a bottleneck. Flows that share a bottleneck
link usually compete with one another for their share of the link usually compete with one another for their share of the
capacity. This competition has the potential to increase packet loss capacity. This competition has the potential to increase packet loss
and delays. This is especially relevant for interactive applications and delays. This is especially relevant for interactive applications
skipping to change at page 9, line 27 skipping to change at page 9, line 27
and H2, neither H1 nor H2 alone obtain enough knowledge to detect and H2, neither H1 nor H2 alone obtain enough knowledge to detect
this shared bottleneck; H3 can however determine it by combining this shared bottleneck; H3 can however determine it by combining
the summary statistics related to H1 and H2, respectively. the summary statistics related to H1 and H2, respectively.
3.1. SBD feedback requirements 3.1. SBD feedback requirements
There are three possible scenarios each with different feedback There are three possible scenarios each with different feedback
requirements: requirements:
1. Both summary statistic calculations and SBD are performed at 1. Both summary statistic calculations and SBD are performed at
senders only. senders only. When sender-based congestion control is
implemented, this method is RECOMMENDED.
2. Summary statistics calculated on the receivers and SBD at the 2. Summary statistics calculated on the receivers and SBD at the
senders. senders.
3. Summary statistic calculations on receivers, and SBD performed at 3. Summary statistic calculations on receivers, and SBD performed at
both senders and receivers (beyond the current scope, but allows both senders and receivers (beyond the current scope, but allows
cooperative detection of bottlenecks). cooperative detection of bottlenecks).
Note that the mechanism bases its calculations on the interval T. It All three possibilities are discussed for completeness in this
does not require T to be the feedback interval, only that document, however, it is expected that feedback will take the form
calculations can be performed over measurements made in that scenario 1 and operate in conjunction with sender-based congestion
interval. control mechanisms.
3.1.1. Feedback when all the logic is placed at the sender 3.1.1. Feedback when all the logic is placed at the sender
Having the sender calculate the summary statistics and determine the Having the sender calculate the summary statistics and determine the
shared bottlenecks based on them has the advantage of placing most of shared bottlenecks based on them has the advantage of placing most of
the functionality in one place -- the sender. the functionality in one place -- the sender.
The sender requires precise accurate OWD measurements for every For every packet, the sender requires accurate OWD measurements of
packet, along with an indication of lost packets (or the proportion adequate precision, along with an indication of lost packets (or the
of packets lost over the interval T). The mechanism performs its proportion of packets lost over an interval). These can be provided
calculations every T and requires measurements to be available for by [I-D.dt-rmcat-feedback-message].
this.
It is expected that the draft-ietf-rmcat-feedback-message will The mechanism performs its calculation whenever it has received
provide the necessary feedback for both SBD and congestion sufficient measurements in the feedback messages to cover the T base
controllers. time interval. The exact timing of these calculations will depend on
the frequency of the feedback message.
3.1.2. Feedback when the statistics are calculated at the receiver and 3.1.2. Feedback when the statistics are calculated at the receiver and
SBD performed at the sender SBD performed at the sender
This scenario minimizes feedback, but requires receivers to send This scenario minimizes feedback, but requires receivers to send
selected summary statistics at an agreed regular interval. We selected summary statistics at an agreed regular interval. We
envisage the following exchange of information to initialize the envisage the following exchange of information to initialize the
system: system:
o An initialization message from the sender to the receiver will o An initialization message from the sender to the receiver will
skipping to change at page 10, line 44 skipping to change at page 10, line 45
* The values of T, N, M, and the necessary resolution and * The values of T, N, M, and the necessary resolution and
precision of the relayed statistics. precision of the relayed statistics.
o A response message from the receiver acknowledges this message o A response message from the receiver acknowledges this message
with a list of key metrics it supports (subset of the senders with a list of key metrics it supports (subset of the senders
list) and is able to relay back to the sender. list) and is able to relay back to the sender.
This initialization exchange may be repeated to finalize the agreed This initialization exchange may be repeated to finalize the agreed
metrics should not all be supported by all receivers. metrics should not all be supported by all receivers.
After initialization the agreed summary statistics will be fed back After initialization the agreed summary statistics SHOULD be fed back
to the sender (nominally every T). to the sender (nominally every T).
3.1.3. Feedback when bottlenecks can be determined at both senders and 3.1.3. Feedback when bottlenecks can be determined at both senders and
receivers receivers
This type of mechanism is currently beyond the scope of SBD in RMCAT. This type of mechanism is currently beyond the scope of SBD in RMCAT.
It is mentioned here to ensure more advanced sender/receiver It is mentioned here to ensure more advanced sender/receiver
cooperative shared bottleneck determination mechanisms remain cooperative shared bottleneck determination mechanisms remain
possible in the future. possible in the future.
skipping to change at page 14, line 39 skipping to change at page 14, line 41
These flows, flows transiting a bottleneck, are then progressively These flows, flows transiting a bottleneck, are then progressively
divided into groups based on the freq_est, var_est, and skew_est divided into groups based on the freq_est, var_est, and skew_est
summary statistics. The process proceeds according to the following summary statistics. The process proceeds according to the following
steps: steps:
2. Group flows whose difference in sorted freq_est is less than a 2. Group flows whose difference in sorted freq_est is less than a
threshold: threshold:
diff(freq_est) < p_f diff(freq_est) < p_f
3. Subdivide freq_est groups by grouping flows whose difference in 3. Subdivide the groups obtained in 2. by grouping flows whose
sorted E_M(var_est) (highest to lowest) is less than a threshold: difference in sorted E_M(var_est) (highest to lowest) is less
than a threshold:
diff(var_est) < (p_mad * var_est) diff(var_est) < (p_mad * var_est)
The threshold, (p_mad * var_est), is with respect to the highest The threshold, (p_mad * var_est), is with respect to the highest
value in the difference. value in the difference.
4. Subdivide var_est groups by grouping flows whose difference in 4. Subdivide the groups obtained in 3. by grouping flows whose
sorted skew_est is less than a threshold: difference in sorted skew_est is less than a threshold:
diff(skew_est) < p_s diff(skew_est) < p_s
5. When packet loss is high enough to be reliable (pkt_loss > p_l), 5. When packet loss is high enough to be reliable (pkt_loss > p_l),
Subdivide skew_est groups by grouping flows whose difference is Subdivide the groups obtained in 4. by grouping flows whose
less than a threshold difference is less than a threshold
diff(pkt_loss) < (p_d * pkt_loss) diff(pkt_loss) < (p_d * pkt_loss)
The threshold, (p_d * pkt_loss), is with respect to the highest The threshold, (p_d * pkt_loss), is with respect to the highest
value in the difference. value in the difference.
This procedure involves sorting estimates from highest to lowest. It This procedure involves sorting estimates from highest to lowest. It
is simple to implement, and efficient for small numbers of flows (up is simple to implement, and efficient for small numbers of flows (up
to 10-20). to 10-20).Figure 2 illustrates this algorithm
*********
* Flows *
***.**.**
/ '
/ '--.
/ \
.---v--. .----v---.
1. Flows traversing | Cong | | UnCong |
a bottleneck '-.--.-' '--------'
/ \
/ \
/ \
.--v--. v-----.
2. Divide by | g_1 | ... | g_n |
freq_est '---.-. '----..
/ \ / \
/ '--. v '------.
/ \ \
.----v-. .-v----. .---v--.
3. Divide by | g_1a | ... | g_1z | ... | g_nz |
var_est '---.-.' '-----.. '-.-.--'
/ \ / \ / |
/ '-----. v v v |
/ \ |
.-v-----. .-v-----. .---v---.
4. Divide by | g_1ai | ... | g_1ax | ... | g_nzx |
skew_est '----.-.' '------.. '-.-.---'
/ \ / \ / |
/ '--. v v v |
/ \ |
.-----v--. .-v------. .----v---.
5. Divide by | g_1aiA | ... | g_1aiZ | ... | g_nzxZ |
pkt_loss '--------' '--------' '--------'
(when applicable)
Simple grouping algorithm.
Figure 2
3.3.2. Using the flow group signal 3.3.2. Using the flow group signal
Grouping decisions can be made every T from the second T, however Grouping decisions can be made every T from the second T, however
they will not attain their full design accuracy until after the they will not attain their full design accuracy until after the
2*N'th T interval. We recommend that grouping decisions are not made 2*N'th T interval. We recommend that grouping decisions are not made
until 2*M T intervals. until 2*M T intervals.
Network conditions, and even the congestion controllers, can cause Network conditions, and even the congestion controllers, can cause
bottlenecks to fluctuate. A coupled congestion controller MAY decide bottlenecks to fluctuate. A coupled congestion controller MAY decide
skipping to change at page 19, line 27 skipping to change at page 21, line 27
clock offsets should be approximately constant over the measurement clock offsets should be approximately constant over the measurement
periods, the offset is subtracted out in the calculation. periods, the offset is subtracted out in the calculation.
5.1. Time stamp resolution 5.1. Time stamp resolution
The SBD mechanism requires timing information precise enough to be The SBD mechanism requires timing information precise enough to be
able to make comparisons. As a rule of thumb, the time resolution able to make comparisons. As a rule of thumb, the time resolution
should be less than one hundredth of a typical path's range of should be less than one hundredth of a typical path's range of
delays. In general, the coarser the time resolution, the more care delays. In general, the coarser the time resolution, the more care
that needs to be taken to ensure rounding errors do not bias the that needs to be taken to ensure rounding errors do not bias the
skewness calculation. skewness calculation. Time stamp resolution such as that described
by [I-D.dt-rmcat-feedback-message] should be sufficient.
Typical RTP media flows use sub-millisecond timers, which should be
adequate in most situations.
5.2. Clock skew 5.2. Clock skew
Generally sender and receiver clock skew will be too small to cause Generally sender and receiver clock skew will be too small to cause
significant errors in the estimators. Skew_est and freq_est are the significant errors in the estimators. Skew_est and freq_est are the
most sensitive to this type of noise due to their use of a mean OWD most sensitive to this type of noise due to their use of a mean OWD
calculated over a longer interval. In circumstances where clock skew calculated over a longer interval. In circumstances where clock skew
is high, basing skew_est only on the previous T's mean and ignoring is high, basing skew_est only on the previous T's mean and ignoring
freq_est provides a noisier but reliable signal. freq_est provides a noisier but reliable signal.
skipping to change at page 20, line 32 skipping to change at page 22, line 30
Non-authenticated RTCP packets carrying OWD measurements, shared Non-authenticated RTCP packets carrying OWD measurements, shared
bottleneck indications, and/or summary statistics could allow bottleneck indications, and/or summary statistics could allow
attackers to alter the bottleneck sharing characteristics for private attackers to alter the bottleneck sharing characteristics for private
gain or disruption of other parties communication. gain or disruption of other parties communication.
10. Change history 10. Change history
Changes made to this document: Changes made to this document:
WG-06->WG-07 : Updates addressing
https://mailarchive.ietf.org/arch/msg/
rmcat/80B6q4nI7carGcf_ddBwx7nKvOw. Mainly
clarifications. Figure 2 to supplement grouping
algorithm description.
WG-05->WG-06 : Updates addressing WG reviews WG-05->WG-06 : Updates addressing WG reviews
https://mailarchive.ietf.org/arch/msg/rmcat/- https://mailarchive.ietf.org/arch/msg/rmcat/-
1JdrTMq1Y5T6ZNlOkrQJQ27TzE and 1JdrTMq1Y5T6ZNlOkrQJQ27TzE and
https://mailarchive.ietf.org/arch/msg/rmcat/ https://mailarchive.ietf.org/arch/msg/rmcat/
eI2Q1f8NL2SxbJgjFLR4_rEmJ_g. This has mainly eI2Q1f8NL2SxbJgjFLR4_rEmJ_g. This has mainly
involved minor clarifications, including the moving involved minor clarifications, including the moving
of 3.4.1 and 3.5 into the new Section 4, and 3.4.1 of 3.4.1 and 3.5 into the new Section 4, and 3.4.1
into Section 5 into Section 5
WG-04->WG-05 : Fix ToC formatting. Add section on expected WG-04->WG-05 : Fix ToC formatting. Add section on expected
skipping to change at page 22, line 13 skipping to change at page 24, line 13
11.2. Informative References 11.2. Informative References
[Hayes-LCN14] [Hayes-LCN14]
Hayes, D., Ferlin, S., and M. Welzl, "Practical Passive Hayes, D., Ferlin, S., and M. Welzl, "Practical Passive
Shared Bottleneck Detection using Shape Summary Shared Bottleneck Detection using Shape Summary
Statistics", Proc. the IEEE Local Computer Networks Statistics", Proc. the IEEE Local Computer Networks
(LCN) pp150-158, September 2014, (LCN) pp150-158, September 2014,
<http://heim.ifi.uio.no/davihay/ <http://heim.ifi.uio.no/davihay/
hayes14__pract_passiv_shared_bottl_detec-abstract.html>. hayes14__pract_passiv_shared_bottl_detec-abstract.html>.
[I-D.dt-rmcat-feedback-message]
Sarker, Z., Perkins, C., Singh, V., and M. Ramalho, "RTP
Control Protocol (RTCP) Feedback for Congestion Control",
draft-dt-rmcat-feedback-message-02 (work in progress), May
2017.
[I-D.ietf-rmcat-coupled-cc] [I-D.ietf-rmcat-coupled-cc]
Islam, S., Welzl, M., and S. Gjessing, "Coupled congestion Islam, S., Welzl, M., and S. Gjessing, "Coupled congestion
control for RTP media", draft-ietf-rmcat-coupled-cc-00 control for RTP media", draft-ietf-rmcat-coupled-cc-06
(work in progress), September 2015. (work in progress), March 2017.
[RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V.
Jacobson, "RTP: A Transport Protocol for Real-Time Jacobson, "RTP: A Transport Protocol for Real-Time
Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550, Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550,
July 2003, <http://www.rfc-editor.org/info/rfc3550>. July 2003, <http://www.rfc-editor.org/info/rfc3550>.
[RFC4585] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey, [RFC4585] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey,
"Extended RTP Profile for Real-time Transport Control "Extended RTP Profile for Real-time Transport Control
Protocol (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585, Protocol (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585,
DOI 10.17487/RFC4585, July 2006, DOI 10.17487/RFC4585, July 2006,
 End of changes. 19 change blocks. 
50 lines changed or deleted 100 lines changed or added

This html diff was produced by rfcdiff 1.45. The latest version is available from http://tools.ietf.org/tools/rfcdiff/