2 .\" Copyright (c) 2013 Swinburne University of Technology, Melbourne, Australia
3 .\" All rights reserved.
5 .\" Redistribution and use in source and binary forms, with or without
6 .\" modification, are permitted provided that the following conditions
8 .\" 1. Redistributions of source code must retain the above copyright
9 .\" notice, this list of conditions and the following disclaimer.
10 .\" 2. Redistributions in binary form must reproduce the above copyright
11 .\" notice, this list of conditions and the following disclaimer in the
12 .\" documentation and/or other materials provided with the distribution.
14 .\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
18 .\" ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33 .Nd CDG Congestion Control Algorithm
35 CAIA-Delay Gradient (CDG) is a hybrid congestion control algorithm which reacts
36 to both packet loss and inferred queuing delay.
37 It attempts to operate as a delay-based algorithm where possible, but utilises
38 heuristics to detect loss-based TCP cross traffic and will compete effectively
40 CDG is therefore incrementally deployable and suitable for use on shared
43 During delay-based operation, CDG uses a delay-gradient based probabilistic
44 backoff mechanism, and will also try to infer non congestion related
45 packet losses and avoid backing off when they occur.
46 During loss-based operation, CDG essentially reverts to
47 .Xr cc_newreno 4 Ns - Ns like
50 CDG switches to loss-based operation when it detects that a configurable number
51 of consecutive delay-based backoffs have had no measurable effect.
52 It periodically attempts to return to delay-based operation, but will keep
53 switching back to loss-based operation as required.
55 The algorithm exposes the following variables in the
56 .Va net.inet.tcp.cc.cdg
60 .Bl -tag -width ".Va exp_backoff_scale"
62 Current algorithm/implementation version number.
64 Delay-based window decrease factor as a percentage (on delay-based backoff, w =
65 w * beta_delay / 100).
68 Loss-based window decrease factor as a percentage (on loss-based backoff, w =
71 .It Va exp_backoff_scale
72 Scaling parameter for the probabilistic exponential backoff.
74 .It Va smoothing_factor
75 Number of samples used for moving average smoothing (0 means no smoothing).
77 .It Va loss_compete_consec_cong
78 Number of consecutive delay-gradient based congestion episodes which will
79 trigger loss-based CC compatibility.
81 .It Va loss_compete_hold_backoff
82 Number of consecutive delay-gradient based congestion episodes to hold the
83 window backoff for loss-based CC compatibility.
86 If non-zero, this enables an experimental mode where CDG's window increase
87 factor (alpha) is increased by 1 MSS every
89 RTTs during congestion avoidance mode.
92 to 1 results in the most aggressive growth of the window increase factor over
96 values for slower growth.)
114 .%T "Revisiting TCP Congestion Control using Delay Gradients"
115 .%J "Networking 2011 Proceedings, Part II"
122 .%T "Minimising RTT across homogeneous 802.11 WLANs with CAIA Delay-Gradient TCP (v0.1)"
123 .%R "CAIA Technical Report 121113A"
125 .%U "http://caia.swin.edu.au/reports/121113A/CAIA-TR-121113A.pdf"
128 Development and testing of this software were made possible in part by grants
129 from the FreeBSD Foundation and The Cisco University Research Program Fund, a
130 corporate advised fund of Silicon Valley Community Foundation.
134 congestion control module first appeared in
137 The module was first released in 2011 by David Hayes whilst working on the
138 NewTCP research project at Swinburne University of Technology's Centre for
139 Advanced Internet Architectures, Melbourne, Australia.
140 More details are available at:
142 http://caia.swin.edu.au/urp/newtcp/
147 congestion control module was written by
148 .An David Hayes Aq david.hayes@ieee.org .
149 This manual page was written by
150 .An Lawrence Stewart Aq lstewart@FreeBSD.org
152 .An Grenville Armitage Aq garmitage@swin.edu.au .
154 The underlying algorithm and parameter values are still a work in progress and
155 may not be optimal for some network scenarios.