xref: /freebsd/share/man/man4/cc_vegas.4 (revision d37eb51047221dc3322b34db1038ff3aa533883f)
1.\"
2.\" Copyright (c) 2010-2011 The FreeBSD Foundation
3.\"
4.\" This documentation was written at the Centre for Advanced Internet
5.\" Architectures, Swinburne University of Technology, Melbourne, Australia by
6.\" David Hayes under sponsorship from the FreeBSD Foundation.
7.\"
8.\" Redistribution and use in source and binary forms, with or without
9.\" modification, are permitted provided that the following conditions
10.\" are met:
11.\" 1. Redistributions of source code must retain the above copyright
12.\"    notice, this list of conditions and the following disclaimer.
13.\" 2. Redistributions in binary form must reproduce the above copyright
14.\"    notice, this list of conditions and the following disclaimer in the
15.\"    documentation and/or other materials provided with the distribution.
16.\"
17.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
21.\" ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27.\" SUCH DAMAGE.
28.\"
29.Dd May 13, 2021
30.Dt CC_VEGAS 4
31.Os
32.Sh NAME
33.Nm cc_vegas
34.Nd Vegas Congestion Control Algorithm
35.Sh DESCRIPTION
36The Vegas congestion control algorithm uses what the authors term the actual and
37expected transmission rates to determine whether there is congestion along the
38network path i.e.
39.Bl -item -offset indent
40.It
41actual rate = (total data sent in a RTT) / RTT
42.It
43expected rate = cwnd / RTTmin
44.It
45diff = expected - actual
46.El
47.Pp
48where RTT is the measured instantaneous round trip time and RTTmin is the
49smallest round trip time observed during the connection.
50.Pp
51The algorithm aims to keep diff between two parameters alpha and beta, such
52that:
53.Bl -item -offset indent
54.It
55alpha < diff < beta
56.El
57.Pp
58If diff > beta, congestion is inferred and cwnd is decremented by one packet (or
59the maximum TCP segment size).
60If diff < alpha, then cwnd is incremented by one packet.
61Alpha and beta govern the amount of buffering along the path.
62.Pp
63The implementation was done in a clean-room fashion, and is based on the
64paper referenced in the
65.Sx SEE ALSO
66section below.
67.Sh IMPLEMENTATION NOTES
68The time from the transmission of a marked packet until the receipt of an
69acknowledgement for that packet is measured once per RTT.
70This implementation does not implement Brakmo's and Peterson's original
71duplicate ACK policy since clock ticks in today's machines are not as coarse as
72they were (i.e. 500ms) when Vegas was originally designed.
73Note that modern TCP recovery processes such as fast retransmit and SACK are
74enabled by default in the TCP stack.
75.Sh MIB Variables
76The algorithm exposes the following tunable variables in the
77.Va net.inet.tcp.cc.vegas
78branch of the
79.Xr sysctl 3
80MIB:
81.Bl -tag -width ".Va alpha"
82.It Va alpha
83Query or set the Vegas alpha parameter as a number of buffers on the path.
84When setting alpha, the value must satisfy: 0 < alpha < beta.
85Default is 1.
86.It Va beta
87Query or set the Vegas beta parameter as a number of buffers on the path.
88When setting beta, the value must satisfy: 0 < alpha < beta.
89Default is 3.
90.El
91.Sh SEE ALSO
92.Xr cc_cdg 4 ,
93.Xr cc_chd 4 ,
94.Xr cc_cubic 4 ,
95.Xr cc_dctcp 4 ,
96.Xr cc_hd 4 ,
97.Xr cc_htcp 4 ,
98.Xr cc_newreno 4 ,
99.Xr h_ertt 4 ,
100.Xr mod_cc 4 ,
101.Xr tcp 4 ,
102.Xr khelp 9 ,
103.Xr mod_cc 9
104.Rs
105.%A "L. S. Brakmo"
106.%A "L. L. Peterson"
107.%T "TCP Vegas: end to end congestion avoidance on a global internet"
108.%J "IEEE J. Sel. Areas Commun."
109.%D "October 1995"
110.%V "13"
111.%N "8"
112.%P "1465-1480"
113.Re
114.Sh ACKNOWLEDGEMENTS
115Development and testing of this software were made possible in part by grants
116from the FreeBSD Foundation and Cisco University Research Program Fund at
117Community Foundation Silicon Valley.
118.Sh HISTORY
119The
120.Nm
121congestion control module first appeared in
122.Fx 9.0 .
123.Pp
124The module was first released in 2010 by David Hayes whilst working on the
125NewTCP research project at Swinburne University of Technology's Centre for
126Advanced Internet Architectures, Melbourne, Australia.
127More details are available at:
128.Pp
129http://caia.swin.edu.au/urp/newtcp/
130.Sh AUTHORS
131.An -nosplit
132The
133.Nm
134congestion control module and this manual page were written by
135.An David Hayes Aq Mt david.hayes@ieee.org .
136