1.\" 2.\" Copyright (c) 2013 Swinburne University of Technology, Melbourne, Australia 3.\" All rights reserved. 4.\" 5.\" Redistribution and use in source and binary forms, with or without 6.\" modification, are permitted provided that the following conditions 7.\" are met: 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. 13.\" 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 24.\" SUCH DAMAGE. 25.\" 26.\" $FreeBSD$ 27.\" 28.Dd May 13, 2021 29.Dt CC_CDG 4 30.Os 31.Sh NAME 32.Nm cc_cdg 33.Nd CDG Congestion Control Algorithm 34.Sh DESCRIPTION 35CAIA-Delay Gradient (CDG) is a hybrid congestion control algorithm which reacts 36to both packet loss and inferred queuing delay. 37It attempts to operate as a delay-based algorithm where possible, but utilises 38heuristics to detect loss-based TCP cross traffic and will compete effectively 39as required. 40CDG is therefore incrementally deployable and suitable for use on shared 41networks. 42.Pp 43During delay-based operation, CDG uses a delay-gradient based probabilistic 44backoff mechanism, and will also try to infer non congestion related 45packet losses and avoid backing off when they occur. 46During loss-based operation, CDG essentially reverts to 47.Xr cc_newreno 4 Ns - Ns like 48behaviour. 49.Pp 50CDG switches to loss-based operation when it detects that a configurable number 51of consecutive delay-based backoffs have had no measurable effect. 52It periodically attempts to return to delay-based operation, but will keep 53switching back to loss-based operation as required. 54.Sh MIB Variables 55The algorithm exposes the following variables in the 56.Va net.inet.tcp.cc.cdg 57branch of the 58.Xr sysctl 3 59MIB: 60.Bl -tag -width ".Va exp_backoff_scale" 61.It Va version 62Current algorithm/implementation version number. 63.It Va beta_delay 64Delay-based window decrease factor as a percentage (on delay-based backoff, w = 65w * beta_delay / 100). 66Default is 70. 67.It Va beta_loss 68Loss-based window decrease factor as a percentage (on loss-based backoff, w = 69w * beta_loss / 100). 70Default is 50. 71.It Va exp_backoff_scale 72Scaling parameter for the probabilistic exponential backoff. 73Default is 2. 74.It Va smoothing_factor 75Number of samples used for moving average smoothing (0 means no smoothing). 76Default is 8. 77.It Va loss_compete_consec_cong 78Number of consecutive delay-gradient based congestion episodes which will 79trigger loss-based CC compatibility. 80Default is 5. 81.It Va loss_compete_hold_backoff 82Number of consecutive delay-gradient based congestion episodes to hold the 83window backoff for loss-based CC compatibility. 84Default is 5. 85.It Va alpha_inc 86If non-zero, this enables an experimental mode where CDG's window increase 87factor (alpha) is increased by 1 MSS every 88.Va alpha_inc 89RTTs during congestion avoidance mode. 90(Setting 91.Va alpha_inc 92to 1 results in the most aggressive growth of the window increase factor over 93time. 94Use higher 95.Va alpha_inc 96values for slower growth.) 97Default is 0. 98.El 99.Sh SEE ALSO 100.Xr cc_chd 4 , 101.Xr cc_cubic 4 , 102.Xr cc_dctcp 4 , 103.Xr cc_hd 4 , 104.Xr cc_htcp 4 , 105.Xr cc_newreno 4 , 106.Xr cc_vegas 4 , 107.Xr h_ertt 4 , 108.Xr mod_cc 4 , 109.Xr tcp 4 , 110.Xr khelp 9 , 111.Xr mod_cc 9 112.Rs 113.%A "D. A. Hayes" 114.%A "G. Armitage" 115.%T "Revisiting TCP Congestion Control using Delay Gradients" 116.%J "Networking 2011 Proceedings, Part II" 117.%D "May 2011" 118.%P "328-341" 119.Re 120.Rs 121.%A "N. Khademi" 122.%A "G. Armitage" 123.%T "Minimising RTT across homogeneous 802.11 WLANs with CAIA Delay-Gradient TCP (v0.1)" 124.%R "CAIA Technical Report 121113A" 125.%D "November 2012" 126.%U "http://caia.swin.edu.au/reports/121113A/CAIA-TR-121113A.pdf" 127.Re 128.Sh ACKNOWLEDGEMENTS 129Development and testing of this software were made possible in part by grants 130from the FreeBSD Foundation and The Cisco University Research Program Fund, a 131corporate advised fund of Silicon Valley Community Foundation. 132.Sh HISTORY 133The 134.Nm 135congestion control module first appeared in 136.Fx 9.2 . 137.Pp 138The module was first released in 2011 by David Hayes whilst working on the 139NewTCP research project at Swinburne University of Technology's Centre for 140Advanced Internet Architectures, Melbourne, Australia. 141More details are available at: 142.Pp 143http://caia.swin.edu.au/urp/newtcp/ 144.Sh AUTHORS 145.An -nosplit 146The 147.Nm 148congestion control module was written by 149.An David Hayes Aq Mt david.hayes@ieee.org . 150This manual page was written by 151.An Lawrence Stewart Aq Mt lstewart@FreeBSD.org 152and 153.An Grenville Armitage Aq Mt garmitage@swin.edu.au . 154.Sh BUGS 155The underlying algorithm and parameter values are still a work in progress and 156may not be optimal for some network scenarios. 157