draft-singh-rmcat-cc-eval-03.txt   draft-singh-rmcat-cc-eval-04.txt 
RMCAT WG V. Singh RMCAT WG V. Singh
Internet-Draft J. Ott Internet-Draft J. Ott
Intended status: Informational Aalto University Intended status: Informational Aalto University
Expires: January 16, 2014 July 15, 2013 Expires: April 23, 2014 October 20, 2013
Evaluating Congestion Control for Interactive Real-time Media Evaluating Congestion Control for Interactive Real-time Media
draft-singh-rmcat-cc-eval-03.txt draft-singh-rmcat-cc-eval-04
Abstract Abstract
The Real-time Transport Protocol (RTP) is used to transmit media in The Real-time Transport Protocol (RTP) is used to transmit media in
telephony and video conferencing applications. This document telephony and video conferencing applications. This document
describes the guidelines to evaluate new congestion control describes the guidelines to evaluate new congestion control
algorithms for interactive point-to-point real-time media. algorithms for interactive point-to-point real-time media.
Status of This Memo Status of This Memo
skipping to change at page 1, line 33 skipping to change at page 1, line 33
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 January 16, 2014. This Internet-Draft will expire on April 23, 2014.
Copyright Notice Copyright Notice
Copyright (c) 2013 IETF Trust and the persons identified as the Copyright (c) 2013 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 14 skipping to change at page 2, line 14
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3. Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3.1. RTP Log Format . . . . . . . . . . . . . . . . . . . . . 5 3.1. RTP Log Format . . . . . . . . . . . . . . . . . . . . . 5
4. Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . 5 4. Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.1. Avoiding Congestion Collapse . . . . . . . . . . . . . . 5 4.1. Avoiding Congestion Collapse . . . . . . . . . . . . . . 5
4.2. Stability . . . . . . . . . . . . . . . . . . . . . . . . 5 4.2. Stability . . . . . . . . . . . . . . . . . . . . . . . . 5
4.3. Media Traffic . . . . . . . . . . . . . . . . . . . . . . 5 4.3. Media Traffic . . . . . . . . . . . . . . . . . . . . . . 6
4.4. Start-up Behaviour . . . . . . . . . . . . . . . . . . . 6 4.4. Start-up Behaviour . . . . . . . . . . . . . . . . . . . 6
4.5. Diverse Environments . . . . . . . . . . . . . . . . . . 6 4.5. Diverse Environments . . . . . . . . . . . . . . . . . . 6
4.6. Varying Path Characteristics . . . . . . . . . . . . . . 6 4.6. Varying Path Characteristics . . . . . . . . . . . . . . 7
4.7. Reacting to Transient Events or Interruptions . . . . . . 6 4.7. Reacting to Transient Events or Interruptions . . . . . . 7
4.8. Fairness With Similar Cross-Traffic . . . . . . . . . . . 7 4.8. Fairness With Similar Cross-Traffic . . . . . . . . . . . 7
4.9. Impact on Cross-Traffic . . . . . . . . . . . . . . . . . 7 4.9. Impact on Cross-Traffic . . . . . . . . . . . . . . . . . 7
4.10. Extensions to RTP/RTCP . . . . . . . . . . . . . . . . . 7 4.10. Extensions to RTP/RTCP . . . . . . . . . . . . . . . . . 8
5. Minimum Requirements for Evaluation . . . . . . . . . . . . . 7 5. Minimum Requirements for Evaluation . . . . . . . . . . . . . 8
6. Evaluation Parameters . . . . . . . . . . . . . . . . . . . . 7 6. Evaluation Parameters . . . . . . . . . . . . . . . . . . . . 8
6.1. Bottleneck Traffic Flows . . . . . . . . . . . . . . . . 8 6.1. Bottleneck Traffic Flows . . . . . . . . . . . . . . . . 8
6.2. Access Links . . . . . . . . . . . . . . . . . . . . . . 8 6.2. Access Links . . . . . . . . . . . . . . . . . . . . . . 9
6.3. Bottleneck Link Parameters . . . . . . . . . . . . . . . 9 6.3. Example Bottleneck Link Parameters . . . . . . . . . . . 9
6.4. Router Queue Parameters . . . . . . . . . . . . . . . . . 10 6.4. DropTail Router Queue Parameters . . . . . . . . . . . . 10
6.5. Media Flow Parameters . . . . . . . . . . . . . . . . . . 10 6.5. Media Flow Parameters . . . . . . . . . . . . . . . . . . 11
6.6. Cross-traffic Parameters . . . . . . . . . . . . . . . . 11 6.6. Cross-traffic Parameters . . . . . . . . . . . . . . . . 11
7. Status of Proposals . . . . . . . . . . . . . . . . . . . . . 11 7. Status of Proposals . . . . . . . . . . . . . . . . . . . . . 11
8. Security Considerations . . . . . . . . . . . . . . . . . . . 11 8. Security Considerations . . . . . . . . . . . . . . . . . . . 12
9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 11 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 12
10. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 12 10. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 12
11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 12 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 12
12. References . . . . . . . . . . . . . . . . . . . . . . . . . 12 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 12
12.1. Normative References . . . . . . . . . . . . . . . . . . 12 12.1. Normative References . . . . . . . . . . . . . . . . . . 12
12.2. Informative References . . . . . . . . . . . . . . . . . 13 12.2. Informative References . . . . . . . . . . . . . . . . . 13
Appendix A. Proposal to evaluate Self-fairness of RMCAT Appendix A. Application Trade-off . . . . . . . . . . . . . . . 14
congestion control algorithm . . . . . . . . . . . . 13 A.1. Measuring Quality . . . . . . . . . . . . . . . . . . . . 14
A.1. Evaluation Parameters . . . . . . . . . . . . . . . . . . 15 Appendix B. Proposal to evaluate Self-fairness of RMCAT
A.1.1. Media Traffic Generator . . . . . . . . . . . . . . . 15 congestion control algorithm . . . . . . . . . . . . 14
A.1.2. Bottleneck Link Bandwidth . . . . . . . . . . . . . . 15 B.1. Evaluation Parameters . . . . . . . . . . . . . . . . . . 15
A.1.3. Bottleneck Link Queue Type and Length . . . . . . . . 15 B.1.1. Media Traffic Generator . . . . . . . . . . . . . . . 15
A.1.4. RMCAT flows and delay legs . . . . . . . . . . . . . 15 B.1.2. Bottleneck Link Bandwidth . . . . . . . . . . . . . . 16
A.1.5. Impairment Generator . . . . . . . . . . . . . . . . 16 B.1.3. Bottleneck Link Queue Type and Length . . . . . . . . 16
A.2. Proposed Passing Criteria . . . . . . . . . . . . . . . . 16 B.1.4. RMCAT flows and delay legs . . . . . . . . . . . . . 16
A.3. Extensability of the Experiment . . . . . . . . . . . . . 17 B.1.5. Impairment Generator . . . . . . . . . . . . . . . . 17
Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 17 B.2. Proposed Passing Criteria . . . . . . . . . . . . . . . . 17
B.1. Changes in draft-singh-rmcat-cc-eval-03 . . . . . . . . . 17 B.3. Extensibility of the Experiment . . . . . . . . . . . . . 17
B.2. Changes in draft-singh-rmcat-cc-eval-02 . . . . . . . . . 17 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 18
B.3. Changes in draft-singh-rmcat-cc-eval-01 . . . . . . . . . 17 C.1. Changes in draft-singh-rmcat-cc-eval-04 . . . . . . . . . 18
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 18 C.2. Changes in draft-singh-rmcat-cc-eval-03 . . . . . . . . . 18
C.3. Changes in draft-singh-rmcat-cc-eval-02 . . . . . . . . . 18
C.4. Changes in draft-singh-rmcat-cc-eval-01 . . . . . . . . . 18
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 19
1. Introduction 1. Introduction
This memo describes the guidelines to help with evaluating new This memo describes the guidelines to help with evaluating new
congestion control algorithms for interactive point-to-point real congestion control algorithms for interactive point-to-point real
time media. The requirements for the congestion control algorithm time media. The requirements for the congestion control algorithm
are outlined in [I-D.jesup-rmcat-reqs]). This document builds upon are outlined in [I-D.jesup-rmcat-reqs]). This document builds upon
previous work at the IETF: Specifying New Congestion Control previous work at the IETF: Specifying New Congestion Control
Algorithms [RFC5033] and Metrics for the Evaluation of Congestion Algorithms [RFC5033] and Metrics for the Evaluation of Congestion
Control Algorithms [RFC5166]. Control Algorithms [RFC5166].
skipping to change at page 3, line 38 skipping to change at page 3, line 41
The terminology defined in RTP [RFC3550], RTP Profile for Audio and The terminology defined in RTP [RFC3550], RTP Profile for Audio and
Video Conferences with Minimal Control [RFC3551], RTCP Extended Video Conferences with Minimal Control [RFC3551], RTCP Extended
Report (XR) [RFC3611], Extended RTP Profile for RTCP-based Feedback Report (XR) [RFC3611], Extended RTP Profile for RTCP-based Feedback
(RTP/AVPF) [RFC4585] and Support for Reduced-Size RTCP [RFC5506] (RTP/AVPF) [RFC4585] and Support for Reduced-Size RTCP [RFC5506]
apply. apply.
3. Metrics 3. Metrics
[RFC5166] describes the basic metrics for congestion control. [RFC5166] describes the basic metrics for congestion control.
Metrics that are important to interactive multimedia are: Metrics that are of interest for interactive multimedia are:
o Throughput. o Throughput.
o Minimizing oscillations in the transmission rate (stability) when o Minimizing oscillations in the transmission rate (stability) when
the end-to-end capacity varies slowly. the end-to-end capacity varies slowly.
o Delay. o Delay.
o Reactivity to transient events. o Reactivity to transient events.
o Packet losses and discards. o Packet losses and discards.
Each experiment logs every incoming and outgoing packet (the RTP o Section 2.1 of [RFC5166] discusses the tradeoff between
logging format is described in Section 3.1). The logging can be done throughput, delay and loss.
inside the application or at the endpoints using pcap (packet
capture, e.g., tcpdump, wireshark). The following are calculated Each experiment is expected to log every incoming and outgoing packet
based on the information in the packet logs: (the RTP logging format is described in Section 3.1). The logging
can be done inside the application or at the endpoints using pcap
(packet capture, e.g., tcpdump, wireshark). The following are
calculated based on the information in the packet logs:
1. Sending rate, Receiver rate, Goodput 1. Sending rate, Receiver rate, Goodput
2. Packet delay 2. Packet delay
3. Packet loss 3. Packet loss
4. Packets discarded from the playout or de-jitter buffer 4. If using, retransmission or FEC: residual loss
[Editor's note: How to handle packet re-transmissions? loss before 5. Packets discarded from the playout or de-jitter buffer
retransmission, after retransmission?]
[Open issue (1): Instead of defining fairness, there has been [Open issue (1): The "unfairness" test is (measured at 1s intervals):
discussion on defining "unfairness". The criteria are:
1. Do not trigger the circuit breaker. 1. Do not trigger the circuit breaker.
2. Over 3 times or less than 1/3 times the throughput for an RMCAT 2. Over 3 times or less than 1/3 times the throughput for an RMCAT
media stream compared to identical RMCAT streams competing on a media stream compared to identical RMCAT streams competing on a
bottleneck, for a case when the competing streams have similar RTTs. bottleneck, for a case when the competing streams have similar RTTs.
3. Over 3 times delay compared to RTT measurements performed before 3. Over 3 times delay compared to RTT measurements performed before
starting the RMCAT flow or for the case when competing with identical starting the RMCAT flow or for the case when competing with identical
RMCAT streams having similar RTTs. RMCAT streams having similar RTTs.
Here, rather than discussing the number '3'? Does the criteria ]
capture Unfairness adequately?]
[Open issue (2): Convergence time was discussed briefly in the design [Open issue (2): Possibly using Jain-fairness index.]
meetings. It is defined as: the time it takes the congestion control
to reach a stable rate (at startup or after new RMCAT flows are
added). What is a stable rate?]
[Open issue (3): previous versions of the document had Bandwidth Convergence time: the time taken to reach a stable rate at startup,
Utilization, defined as ratio of sending rate to the available after the available link capacity changes, or when new flows get
bottleneck capacity. This is useful when the RMCAT flow is by itself added to the bottleneck link.
or competing with similar flows (where the assumption would be that
all flows get an equal share). Remove this?] Bandwidth Utilization, defined as ratio of the instantaneous sending
rate to the instantaneous bottleneck capacity. This metric is useful
when an RMCAT flow is by itself or competing with similar cross-
traffic.
From the logs the statistical measures (min, max, mean, standard From the logs the statistical measures (min, max, mean, standard
deviation and variance) for the whole duration or any specific part deviation and variance) for the whole duration or any specific part
of the session can be calculated. Also the metrics (sending rate, of the session can be calculated. Also the metrics (sending rate,
receiver rate, goodput, latency) can be visualized in graphs as receiver rate, goodput, latency) can be visualized in graphs as
variation over time, the measurements in the plot are at 1 second variation over time, the measurements in the plot are at 1 second
intervals. Additionally, from the logs it is possible to plot the intervals. Additionally, from the logs it is possible to plot the
histogram or CDF of packet delay. histogram or CDF of packet delay.
Section 2.1 of [RFC5166] discusses the tradeoff between throughput,
delay and loss.
[Open issue (4): Application trade-off is yet to be defined. see
RMCAT requirements [I-D.jesup-rmcat-reqs] document. Perhaps each
experiment should define the application's expectation or trade-off.]
3.1. RTP Log Format 3.1. RTP Log Format
The log file is tab or comma separated containing the following The log file is tab or comma separated containing the following
details: details:
Send or receive timestamp (unix) Send or receive timestamp (unix)
RTP payload type RTP payload type
SSRC SSRC
RTP sequence no RTP sequence no
RTP timestamp RTP timestamp
marker bit marker bit
payload size payload size
[Open issue (5): Should the retransmissions for post-repair loss If the congestion control implements, retransmissions or FEC, the
metric be logged in a separate file? the repair streams have evaluation should report both packet loss (before applying error-
different payload type and/or SSRC.] resilience) and residual packet loss (after applying error-
resilience).
4. Guidelines 4. Guidelines
A congestion control algorithm should be tested in simulation or a A congestion control algorithm should be tested in simulation or a
testbed environment, and the experiments should be repeated multiple testbed environment, and the experiments should be repeated multiple
times to infer statistical significance. The following guidelines times to infer statistical significance. The following guidelines
are considered for evaluation: are considered for evaluation:
4.1. Avoiding Congestion Collapse 4.1. Avoiding Congestion Collapse
The congestion control algorithm is expected to take an action, such
as reducing the sending rate, when it detects congestion. Typically,
it should intervene before the circuit breaker
[I-D.ietf-avtcore-rtp-circuit-breakers] is engaged.
Does the congestion control propose any changes to (or diverge from) Does the congestion control propose any changes to (or diverge from)
the circuit breaker conditions defined in the circuit breaker conditions defined in
[I-D.ietf-avtcore-rtp-circuit-breakers]. [I-D.ietf-avtcore-rtp-circuit-breakers].
4.2. Stability 4.2. Stability
The congestion control should be assessed for its stability when the The congestion control should be assessed for its stability when the
path characteristics do not change over time. Changing the media path characteristics do not change over time. Changing the media
encoding rate estimate too often or by too much may adversely affect encoding rate estimate too often or by too much may adversely affect
the application layer performance. the application layer performance.
4.3. Media Traffic 4.3. Media Traffic
The congestion control algorithm should be assessed with different The congestion control algorithm should be assessed with different
types of media behavior, i.e., the media should contain idle and types of media behavior, i.e., the media should contain idle and
data-limited periods. For example, periods of silence for audio or data-limited periods. For example, periods of silence for audio,
varying amount of motion for video. However, the evaluation can be varying amount of motion for video, or bursty nature of I-frames.
done in two stages. In the first stage, media stream can generate
traffic at the rate calculated by the congestion controller. In the The evaluation may be done in two stages. In the first stage, the
second stage, real codecs or models of video codecs should be used to endpoint generates traffic at the rate calculated by the congestion
mimic real-world cases. controller. In the second stage, real codecs or models of video
codecs are used to mimic application-limited data periods and varying
video frame sizes.
4.4. Start-up Behaviour 4.4. Start-up Behaviour
The congestion control algorithm should be assessed with different The congestion control algorithm should be assessed with different
start-rates. The main reason is to observe the behavior of the start-rates. The main reason is to observe the behavior of the
congestion control in different evaluation scenarios, such as when congestion control in different evaluation scenarios, such as when
competing with varying amount of cross-traffic or how quickly does competing with varying amount of cross-traffic or how quickly does
the congestion control algorithm achieve a stable sending rate. the congestion control algorithm achieve a stable sending rate.
[Editor's note: requires a robust definition for unfriendliness and [Editor's note: requires a robust definition for unfriendliness and
skipping to change at page 6, line 38 skipping to change at page 7, line 9
algorithms may incorrectly identify transmission loss as congestion algorithms may incorrectly identify transmission loss as congestion
loss and reduce the media encoding rate by too much, which may cause loss and reduce the media encoding rate by too much, which may cause
oscillatory behavior and deteriorate the users' quality of oscillatory behavior and deteriorate the users' quality of
experience. Furthermore, packet loss may induce additional delay in experience. Furthermore, packet loss may induce additional delay in
networks with wireless paths due to link-layer retransmissions. networks with wireless paths due to link-layer retransmissions.
4.6. Varying Path Characteristics 4.6. Varying Path Characteristics
The congestion control algorithm should be evaluated for a range of The congestion control algorithm should be evaluated for a range of
path characteristics such as, different end-to-end capacity and path characteristics such as, different end-to-end capacity and
latency, varying amount of cross traffic on a bottle-neck link and a latency, varying amount of cross traffic on a bottleneck link and a
router's queue length. In an experiment, if the media only flows in router's queue length. For the moment, only DropTail queues are
a single direction, the feedback path should also be tested with used. However, if new Active Queue Management (AQM) schemes become
varying amounts of impairments. available, the performance of the congestion control algorithm should
be again evaluated.
In an experiment, if the media only flows in a single direction, the
feedback path should also be tested with varying amounts of
impairments.
The main motivation for the previous and current criteria is to The main motivation for the previous and current criteria is to
identify situations in which the proposed congestion control is less identify situations in which the proposed congestion control is less
performant. performant.
[Open issue (6): Different types of queueing mechanisms? Random
Early Detection or only DropTail?].
4.7. Reacting to Transient Events or Interruptions 4.7. Reacting to Transient Events or Interruptions
The congestion control algorithm should be able to handle changes in The congestion control algorithm should be able to handle changes in
end-to-end capacity and latency. Latency may change due to route end-to-end capacity and latency. Latency may change due to route
updates, link failures, handovers etc. In mobile environment the updates, link failures, handovers etc. In mobile environment the
end-to-end capacity may vary due to the interference, fading, end-to-end capacity may vary due to the interference, fading,
handovers, etc. In wired networks the end-to-end capacity may vary handovers, etc. In wired networks the end-to-end capacity may vary
due to changes in resource reservation. due to changes in resource reservation.
4.8. Fairness With Similar Cross-Traffic 4.8. Fairness With Similar Cross-Traffic
The congestion control algorithm should be evaluated when competing The congestion control algorithm should be evaluated when competing
skipping to change at page 7, line 42 skipping to change at page 8, line 14
4.10. Extensions to RTP/RTCP 4.10. Extensions to RTP/RTCP
The congestion control algorithm should indicate if any protocol The congestion control algorithm should indicate if any protocol
extensions are required to implement it and should carefully describe extensions are required to implement it and should carefully describe
the impact of the extension. the impact of the extension.
5. Minimum Requirements for Evaluation 5. Minimum Requirements for Evaluation
[Editor's Note: If needed, a minimum evaluation criteria can be based [Editor's Note: If needed, a minimum evaluation criteria can be based
on the above guidelines] on the above guidelines or defined tests/scenarios.]
6. Evaluation Parameters 6. Evaluation Parameters
An evaluation scenario is created from a list of network, link and An evaluation scenario is created from a list of network, link and
flow characteristics. The parameters discussed in the following flow characteristics. The example parameters discussed in the
subsections are meant to aid in creating evaluation scenarios and do following subsections are meant to aid in creating evaluation
not describe an evaluation scenario. The scenario discussed in scenarios and do not describe an evaluation scenario. The scenario
Appendix A takes into account all these parameters. discussed in Appendix B takes into account all these parameters.
6.1. Bottleneck Traffic Flows 6.1. Bottleneck Traffic Flows
The network scenario describes the types of flows sharing the common The network scenario describes the types of flows sharing the common
bottleneck with a single RMCAT flow, they are: bottleneck with a single RMCAT flow, they are:
1. A single RMCAT flow by itself. 1. A single RMCAT flow by itself.
2. Competing with similar RMCAT flows. These competing flows may 2. Competing with similar RMCAT flows. These competing flows may
use the same algorithm or another candidate RMCAT algorithm. use the same algorithm or another candidate RMCAT algorithm.
skipping to change at page 8, line 47 skipping to change at page 9, line 17
+---+ | |<------------------------------| | +---+ +---+ | |<------------------------------| | +---+
+-----+ Link +-----+ +-----+ Link +-----+
(...) // \\ (...) (...) // \\ (...)
// \\ // \\
+---+ // \\ +---+ +---+ // \\ +---+
|Sn |====== / \ ======|Rn | |Sn |====== / \ ======|Rn |
+---+ +---+ +---+ +---+
Figure 1: Simple Topology Figure 1: Simple Topology
[Open Issue (7): Discuss more complex topologies] [Open Issue: Discuss more complex topologies]
6.2. Access Links 6.2. Access Links
The media senders and receivers are typically connected to the The media senders and receivers are typically connected to the
bottleneck link, common access links are: bottleneck link, common access links are:
1. Ethernet (LAN) 1. Ethernet (LAN)
2. Wireless LAN (WLAN) 2. Wireless LAN (WLAN)
3. 3G/LTE 3. 3G/LTE
[Open issue (8): need to describe parameters or traces to model WLAN [Open issue: point to a reference containing parameters or traces to
and 3G/LTE.] model WLAN and 3G/LTE.]
A real-world network typically consists of a mixture of links, the A real-world network typically consists of a mixture of links, the
most important aspect is to identify the location of the bottleneck most important aspect is to identify the location of the bottleneck
link. The bottleneck link can move from one node to another link. The bottleneck link can move from one node to another
depending on the amount of cross-traffic or due to the varying link depending on the amount of cross-traffic or due to the varying link
capacity. The design of the experiments should take this into capacity. The design of the experiments should take this into
account. In the simplest case the access link may not be the account. In the simplest case the access link may not be the
bottleneck link but an intermediate node. bottleneck link but an intermediate node.
6.3. Bottleneck Link Parameters 6.3. Example Bottleneck Link Parameters
The bottleneck link carries multiple flows, these flows may be other The bottleneck link carries multiple flows, these flows may be other
RMCAT flows or other types of cross-traffic. The experiments should RMCAT flows or other types of cross-traffic. The experiments should
dimension the bottleneck link based on the number of flows and the dimension the bottleneck link based on the number of flows and the
expected behavior. For example, if 5 media flows are expected to expected behavior. For example, if 5 media flows are expected to
share the bottleneck link equally, the bottleneck link is set to 5 share the bottleneck link equally, the bottleneck link is set to 5
times the desired transmission rate. times the desired transmission rate.
If the experiment carries only media in one direction, then the If the experiment carries only media in one direction, then the
upstream (sender to receiver) bottleneck link carries media packets upstream (sender to receiver) bottleneck link carries media packets
skipping to change at page 10, line 4 skipping to change at page 10, line 23
links the experiment should describe if the links add latency or not. links the experiment should describe if the links add latency or not.
It is possible for experiments to have multiple hops with different It is possible for experiments to have multiple hops with different
link latencies. Experiments are expected to verify that the link latencies. Experiments are expected to verify that the
congestion control is able to work in challenging situations, for congestion control is able to work in challenging situations, for
example over trans-continental and/or satellite links. The example over trans-continental and/or satellite links. The
experiment should pick link latency values from the following: experiment should pick link latency values from the following:
1. Very low latency: 0-1ms 1. Very low latency: 0-1ms
2. Low latency: 50ms 2. Low latency: 50ms
3. High latency: 150ms 3. High latency: 150ms
4. Extreme latency: 300ms 4. Extreme latency: 300ms
[Editor's note: currently describes the latency for a single link,
instead of end-to-end delay. Which is preferred? or both?]
Similarly, to model lossy links, the experiments can choose one of Similarly, to model lossy links, the experiments can choose one of
the following loss rates, the fractional loss is the ratio of packets the following loss rates, the fractional loss is the ratio of packets
lost and packets sent. lost and packets sent.
1. no loss: 0% 1. no loss: 0%
2. 1% 2. 1%
3. 5% 3. 5%
4. 10% 4. 10%
5. 20% 5. 20%
[Open issue (10): how is the loss generated? using traces, Gilbert- These fractional losses can be generated using traces, Gilbert-Elliot
Elliot model, randomly (uncorrelated) loss.] model, randomly (uncorrelated) loss.
6.4. Router Queue Parameters 6.4. DropTail Router Queue Parameters
The router queue length is measured as the time taken to drain the The router queue length is measured as the time taken to drain the
FIFO queue, they are: FIFO queue, they are:
1. QoS-aware (or short): 70ms 1. QoS-aware (or short): 70ms
2. Nominal: 500ms 2. Nominal: 500ms
3. Buffer-bloated: 2000ms 3. Buffer-bloated: 2000ms
However, the size of the queue is typically measured in bytes or However, the size of the queue is typically measured in bytes or
packets and to convert the queue length measured in seconds to queue packets and to convert the queue length measured in seconds to queue
length in bytes: length in bytes:
QueueSize (in bytes) = QueueSize (in sec) x Throughput (in bps)/8 QueueSize (in bytes) = QueueSize (in sec) x Throughput (in bps)/8
skipping to change at page 10, line 45 skipping to change at page 11, line 14
2. Nominal: 500ms 2. Nominal: 500ms
3. Buffer-bloated: 2000ms 3. Buffer-bloated: 2000ms
However, the size of the queue is typically measured in bytes or However, the size of the queue is typically measured in bytes or
packets and to convert the queue length measured in seconds to queue packets and to convert the queue length measured in seconds to queue
length in bytes: length in bytes:
QueueSize (in bytes) = QueueSize (in sec) x Throughput (in bps)/8 QueueSize (in bytes) = QueueSize (in sec) x Throughput (in bps)/8
[Open issue (11): Confirm the above values, do we need to define
parameters for other types of queues?]
6.5. Media Flow Parameters 6.5. Media Flow Parameters
The media sources can be modeled in two ways. In the first, the The media sources can be modeled in two ways. In the first, the
sources always have data to send, i.e., have no data limited sources always have data to send, i.e., have no data limited
intervals and are able to generate the media rate requested by the intervals and are able to generate the media rate requested by the
RMCAT congestion control algorithm. In the second, the traffic RMCAT congestion control algorithm. In the second, the traffic
generator models the behavior of a media codec, mainly the burstiness generator models the behavior of a media codec, mainly the burstiness
(time-varying data produced by a video GOP). (time-varying data produced by a video GOP).
At the beginning of the session, the media sources are configured to At the beginning of the session, the media sources are configured to
skipping to change at page 11, line 22 skipping to change at page 11, line 36
1. 200 kbps 1. 200 kbps
2. 800 kbps 2. 800 kbps
3. 1300 kbps 3. 1300 kbps
4. 4000 kbps 4. 4000 kbps
6.6. Cross-traffic Parameters 6.6. Cross-traffic Parameters
[Open issue(12): TCP cross-traffic parameters are still TBD, mainly Long-lived TCP flows will download data throughout the session and
the bursty TCP. Long-lived TCP flows will download data throughout are expected to have infinite amount of data to send or receive.]
the session and are expected to have infinite amount of data to send
or receive.] [Open issue: short-lived/bursty TCP cross-traffic parameters are
still TBD.
7. Status of Proposals 7. Status of Proposals
Congestion control algorithms are expected to be published as Congestion control algorithms are expected to be published as
"Experimental" documents until they are shown to be safe to deploy. "Experimental" documents until they are shown to be safe to deploy.
An algorithm published as a draft should be experimented in An algorithm published as a draft should be experimented in
simulation, or a controlled environment (testbed) to show its simulation, or a controlled environment (testbed) to show its
applicability. Every congestion control algorithm should include a applicability. Every congestion control algorithm should include a
note describing the environments in which the algorithm is tested and note describing the environments in which the algorithm is tested and
safe to deploy. It is possible that an algorithm is not recommended safe to deploy. It is possible that an algorithm is not recommended
skipping to change at page 12, line 4 skipping to change at page 12, line 18
not safe for deployment but are proposals to experiment with in not safe for deployment but are proposals to experiment with in
simulation/testbeds. While Experimental algorithms are ones that are simulation/testbeds. While Experimental algorithms are ones that are
deemed safe in some environments but require a more thorough deemed safe in some environments but require a more thorough
evaluation (from the community).] evaluation (from the community).]
8. Security Considerations 8. Security Considerations
Security issues have not been discussed in this memo. Security issues have not been discussed in this memo.
9. IANA Considerations 9. IANA Considerations
There are no IANA impacts in this memo. There are no IANA impacts in this memo.
10. Contributors 10. Contributors
The content and concepts within this document are a product of the The content and concepts within this document are a product of the
discussion carried out in the Design Team. discussion carried out in the Design Team.
Michael Ramalho provided the text for the scenario discussed in Michael Ramalho provided the text for the scenario discussed in
Appendix A. Appendix B.
11. Acknowledgements 11. Acknowledgements
Much of this document is derived from previous work on congestion Much of this document is derived from previous work on congestion
control at the IETF. control at the IETF.
The authors would like to thank Harald Alvestrand, Luca De Cicco, The authors would like to thank Harald Alvestrand, Luca De Cicco,
Wesley Eddy, Lars Eggert, Kevin Gross, Vinayak Hegde, Stefan Holmer, Wesley Eddy, Lars Eggert, Kevin Gross, Vinayak Hegde, Stefan Holmer,
Randell Jesup, Piers O'Hanlon, Colin Perkins, Michael Ramalho, Randell Jesup, Piers O'Hanlon, Colin Perkins, Michael Ramalho,
Zaheduzzaman Sarker, Timothy B. Terriberry, Michael Welzl, and Mo Zaheduzzaman Sarker, Timothy B. Terriberry, Michael Welzl, and Mo
skipping to change at page 13, line 39 skipping to change at page 14, line 5
[SA4-LR] S4-050560, 3GPP., "Error Patterns for MBMS Streaming over [SA4-LR] S4-050560, 3GPP., "Error Patterns for MBMS Streaming over
UTRAN and GERAN", 3GPP S4-050560, 5 2008. UTRAN and GERAN", 3GPP S4-050560, 5 2008.
[TCP-eval-suite] [TCP-eval-suite]
Lachlan, A., Marcondes, C., Floyd, S., Dunn, L., Guillier, Lachlan, A., Marcondes, C., Floyd, S., Dunn, L., Guillier,
R., Gang, W., Eggert, L., Ha, S., and I. Rhee, "Towards a R., Gang, W., Eggert, L., Ha, S., and I. Rhee, "Towards a
Common TCP Evaluation Suite", Proc. PFLDnet. 2008, August Common TCP Evaluation Suite", Proc. PFLDnet. 2008, August
2008. 2008.
Appendix A. Proposal to evaluate Self-fairness of RMCAT congestion Appendix A. Application Trade-off
Application trade-off is yet to be defined. see RMCAT requirements
[I-D.jesup-rmcat-reqs] document. Perhaps each experiment should
define the application's expectation or trade-off.
A.1. Measuring Quality
No quality metric is defined for performance evaluation, it is
currently an open issue. However, there is consensus that congestion
control algorithm should be able to show that it is useful for
interactive video by performing analysis using a real codec and video
sequences.
Appendix B. Proposal to evaluate Self-fairness of RMCAT congestion
control algorithm control algorithm
The goal of the experiment discussed in this section is to initially The goal of the experiment discussed in this section is to initially
take out as many unknowns from the scenario. Later experiments can take out as many unknowns from the scenario. Later experiments can
define more complex environments, topologies and media behavior. define more complex environments, topologies and media behavior.
This experiment evaluates the performance of the RMCAT sender This experiment evaluates the performance of the RMCAT sender
competing with other similar RMCAT flows (running the same algorithm competing with other similar RMCAT flows (running the same algorithm
or other RMCAT proposals) on the bottleneck link. There are up to 20 or other RMCAT proposals) on the bottleneck link. There are up to 20
RMCAT flows competing for capacity, but the media only flows in one RMCAT flows competing for capacity, but the media only flows in one
direction, from senders (S1..S20) to receivers (R1..R20) and the direction, from senders (S1..S20) to receivers (R1..R20) and the
skipping to change at page 15, line 5 skipping to change at page 15, line 33
+---+ +---+
Figure 2: Self-fairness Evaluation Setup Figure 2: Self-fairness Evaluation Setup
Loss impairments are applied at Router C and Router D, but only to Loss impairments are applied at Router C and Router D, but only to
the feedback flows. If the losses are set to 0%, it represents a the feedback flows. If the losses are set to 0%, it represents a
case where the return path is over-provisioned for all traffic. In case where the return path is over-provisioned for all traffic. In
later experiments the loss impairments can be added to the media path later experiments the loss impairments can be added to the media path
as well. as well.
A.1. Evaluation Parameters The media sources are configured to send infinite amount of data,
i.e., the sources always have data to send and have no data limited
intervals. Additionally, the media sources are always successful in
generating the media rate requested by the RMCAT congestion control
algorithm. In this experiment, we avoid the potentially complicated
scenario of using media traffic generators that try to model the
behavior of media codecs (mainly the burstiness).
A.1.1. Media Traffic Generator B.1. Evaluation Parameters
B.1.1. Media Traffic Generator
The media source always generates at the rate requested by the The media source always generates at the rate requested by the
congestion control and has infinite data to send. Furthermore, the congestion control and has infinite data to send. Furthermore, the
media packet generator is subject to the following constraints: media packet generator is subject to the following constraints:
1. It MUST emit a packet at least once per 100 ms time interval. 1. It MUST emit a packet at least once per 100 ms time interval.
2. For low media rate source: when generating data at a rate less 2. For low media rate source: when generating data at a rate less
than a maximum length MTU every 100 ms would allow (e.g., 120 than a maximum length MTU every 100 ms would allow (e.g., 120
kbps = 1500 bytes/packet * 10 packets/sec * 8 bits/byte), the kbps = 1500 bytes/packet * 10 packets/sec * 8 bits/byte), the
skipping to change at page 15, line 30 skipping to change at page 16, line 17
rate. rate.
3. For high media rate sources: when generating data at a rate 3. For high media rate sources: when generating data at a rate
greater than a maximum length MTU every 100 ms would allow, the greater than a maximum length MTU every 100 ms would allow, the
source must do so by sending (approximately) maximum MTU sized source must do so by sending (approximately) maximum MTU sized
packets and adjusting the inter-departure interval to be packets and adjusting the inter-departure interval to be
approximately equal. The intent of this to ensure the data is approximately equal. The intent of this to ensure the data is
sent relatively smoothly independent of the bit rate, subject to sent relatively smoothly independent of the bit rate, subject to
the first constraint. the first constraint.
A.1.2. Bottleneck Link Bandwidth B.1.2. Bottleneck Link Bandwidth
The bottleneck link capacity is dimensioned such that each RMCAT flow The bottleneck link capacity is dimensioned such that each RMCAT flow
in an ideal situation with perfectly equal capacity sharing for all in an ideal situation with perfectly equal capacity sharing for all
the flows on the bottleneck obtains the following throughputs: 200 the flows on the bottleneck obtains the following throughputs: 200
kbps, 800 kbps, 1.3 Mbps and 4 Mbps. kbps, 800 kbps, 1.3 Mbps and 4 Mbps.
For example, experiments with five RMCAT flows with an 800 kbps/flow For example, experiments with five RMCAT flows with an 800 kbps/flow
target rate should set the bottleneck link capacity to 4 Mbps. target rate should set the bottleneck link capacity to 4 Mbps.
A.1.3. Bottleneck Link Queue Type and Length B.1.3. Bottleneck Link Queue Type and Length
The bottleneck link queue (Router A) is a simple FIFO queue having a The bottleneck link queue (Router A) is a simple FIFO queue having a
buffer length corresponding to 70 ms, 500 ms or 2000 ms (defined in buffer length corresponding to 70 ms, 500 ms or 2000 ms (defined in
Section 6.4) of delay at the bottleneck link rate (i.e., actual Section 6.4) of delay at the bottleneck link rate (i.e., actual
buffer lengths in bytes are dependent on bottleneck link bandwidth). buffer lengths in bytes are dependent on bottleneck link bandwidth).
A.1.4. RMCAT flows and delay legs B.1.4. RMCAT flows and delay legs
Experiments run with 1, 3, 5, 10 and 20 RMCAT sources, they are Experiments run with 1, 3, 5, 10 and 20 RMCAT sources, they are
outlined as follows: outlined as follows:
1. Experiments with 1, 3, and 5 RMCAT flows, all RMCAT flows 1. Experiments with 1, 3, and 5 RMCAT flows, all RMCAT flows
commence simultaneously. A single delay leg is used and the link commence simultaneously. A single delay leg is used and the link
latency is set to one of the following : 0 ms, 50 ms and 150 ms. latency is set to one of the following : 0 ms, 50 ms and 150 ms.
2. For 10 and 20 source experiments where all RMCAT flows begin 2. For 10 and 20 source experiments where all RMCAT flows begin
simultaneously the sources are split evenly into two different simultaneously the sources are split evenly into two different
skipping to change at page 16, line 35 skipping to change at page 17, line 21
These cases assess if there are any early or late-comer These cases assess if there are any early or late-comer
advantages or disadvantages for a particular algorithm and to see advantages or disadvantages for a particular algorithm and to see
if any unfairness is reproducible or unpredictable. if any unfairness is reproducible or unpredictable.
[Open issue (A.1): which group does the early and late flow belong [Open issue (A.1): which group does the early and late flow belong
to?] to?]
[Open issue (A.2): Start rate for the media flows] [Open issue (A.2): Start rate for the media flows]
A.1.5. Impairment Generator B.1.5. Impairment Generator
Packet loss is created in the reverse path (affects only feedback Packet loss is created in the reverse path (affects only feedback
packets). Cases of 0%, 1%, 5% and 10% are studied for the 1, 3, and packets). Cases of 0%, 1%, 5% and 10% are studied for the 1, 3, and
5 RMCAT flow experiments, losses are not applied to flows with 10 or 5 RMCAT flow experiments, losses are not applied to flows with 10 or
20 RMCAT flows. 20 RMCAT flows.
A.2. Proposed Passing Criteria B.2. Proposed Passing Criteria
[Editor's note: there has been little or no discussion on the below [Editor's note: there has been little or no discussion on the below
criteria, however, they are listed here for the sake of completeness. criteria, however, they are listed here for the sake of completeness.
No unfairness is observed, i.e., at steady state each flow attains a No unfairness is observed, i.e., at steady state each flow attains a
throughput between [ B/(3*N), (3*B)/N ], where B is the link throughput between [ B/(3*N), (3*B)/N ], where B is the link
bandwidth and N is the number of flows. bandwidth and N is the number of flows.
No flow experiences packet loss when queue length is set to 500 ms or No flow experiences packet loss when queue length is set to 500 ms or
greater. greater.
All individual sources must be in their steady state within twenty All individual sources must be in their steady state within twenty
LRTTs (where LRTT is defined as the RTT associated with the flow with LRTTs (where LRTT is defined as the RTT associated with the flow with
the Largest RTT in the experiment). ] the Largest RTT in the experiment). ]
A.3. Extensability of the Experiment B.3. Extensibility of the Experiment
The above scenario describes only RMCAT sources competing for The above scenario describes only RMCAT sources competing for
capacity on the bottleneck link, however, future experiments can use capacity on the bottleneck link, however, future experiments can use
different types of cross-traffic (as described in Section 6.1). different types of cross-traffic (as described in Section 6.1).
Currently, the forward path (carrying media packets) is characterized Currently, the forward path (carrying media packets) is characterized
to add delay and a fixed bottleneck link capacity, in the future to add delay and a fixed bottleneck link capacity, in the future
packet losses and capacity changes can be applied to mimic a wireless packet losses and capacity changes can be applied to mimic a wireless
link layer (for e.g., WiFi, 3G, LTE). Additionally, only losses are link layer (for e.g., WiFi, 3G, LTE). Additionally, only losses are
applied to the reverse path (carrying feedback packets), later applied to the reverse path (carrying feedback packets), later
experiments can apply the same forward path (carrying media packets) experiments can apply the same forward path (carrying media packets)
impairments to the reverse path. impairments to the reverse path.
Appendix B. Change Log Appendix C. Change Log
Note to the RFC-Editor: please remove this section prior to Note to the RFC-Editor: please remove this section prior to
publication as an RFC. publication as an RFC.
B.1. Changes in draft-singh-rmcat-cc-eval-03 C.1. Changes in draft-singh-rmcat-cc-eval-04
o Incorporate feedback from IETF 87, Berlin.
o Clarified metrics: convergence time, bandwidth utilization.
o Changed fairness criteria to fairness test.
o Added measuring pre- and post-repair loss.
o Added open issue of measuring video quality to appendix.
o clarified use of DropTail and AQM.
o Updated text in "Minimum Requirements for Evaluation"
C.2. Changes in draft-singh-rmcat-cc-eval-03
o Incorporate the discussion within the design team. o Incorporate the discussion within the design team.
o Added a section on evaluation parameters, it describes the flow o Added a section on evaluation parameters, it describes the flow
and network characteristics. and network characteristics.
o Added Appendix with self-fairness experiment. o Added Appendix with self-fairness experiment.
B.2. Changes in draft-singh-rmcat-cc-eval-02 o Changed bottleneck parameters from a proposal to an example set.
C.3. Changes in draft-singh-rmcat-cc-eval-02
o Added scenario descriptions. o Added scenario descriptions.
B.3. Changes in draft-singh-rmcat-cc-eval-01 C.4. Changes in draft-singh-rmcat-cc-eval-01
o Removed QoE metrics. o Removed QoE metrics.
o Changed stability to steady-state. o Changed stability to steady-state.
o Added measuring impact against few and many flows. o Added measuring impact against few and many flows.
o Added guideline for idle and data-limited periods. o Added guideline for idle and data-limited periods.
o Added reference to TCP evaluation suite in example evaluation o Added reference to TCP evaluation suite in example evaluation
 End of changes. 52 change blocks. 
113 lines changed or deleted 155 lines changed or added

This html diff was produced by rfcdiff 1.41. The latest version is available from http://tools.ietf.org/tools/rfcdiff/