cc_cubic(4) [freebsd man page]

CC_CUBIC(4)						   BSD Kernel Interfaces Manual 					       CC_CUBIC(4)

NAME

     cc_cubic -- CUBIC Congestion Control Algorithm

DESCRIPTION

     The CUBIC congestion control algorithm was designed to provide increased throughput in fast and long-distance networks.  It attempts to main-
     tain fairness when competing with legacy NewReno TCP in lower speed scenarios where NewReno is able to operate adequately.

     The congestion window is increased as a function of the time elapsed since the last congestion event.  During regular operation, the window
     increase function follows a cubic function, with the inflection point set to be the congestion window value reached at the last congestion
     event.  CUBIC also calculates an estimate of the congestion window that NewReno would have achieved at a given time after a congestion event.
     When updating the congestion window, the algorithm will choose the larger of the calculated CUBIC and estimated NewReno windows.

     CUBIC also backs off less on congestion by changing the multiplicative decrease factor from 1/2 (used by standard NewReno TCP) to 4/5.

     The implementation was done in a clean-room fashion, and is based on the Internet Draft and paper referenced in the SEE ALSO section below.

MIB Variables
     There are currently no tunable MIB variables.

SEE ALSO

     cc_chd(4), cc_hd(4), cc_htcp(4), cc_newreno(4), cc_vegas(4), mod_cc(4), tcp(4), mod_cc(9)

     Sangtae Ha, Injong Rhee, and Lisong Xu, CUBIC for Fast Long-Distance Networks, http://tools.ietf.org/id/draft-rhee-tcpm-cubic-02.txt.

     Sangtae Ha, Injong Rhee, and Lisong Xu, "CUBIC: a new TCP-friendly high-speed TCP variant", SIGOPS Oper. Syst. Rev., 5, 42, 64-74, July 2008.

ACKNOWLEDGEMENTS

     Development and testing of this software were made possible in part by grants from the FreeBSD Foundation and Cisco University Research Pro-
     gram Fund at Community Foundation Silicon Valley.

HISTORY

     The cc_cubic congestion control module first appeared in FreeBSD 9.0.

     The module was first released in 2009 by Lawrence Stewart whilst studying at Swinburne University of Technology's Centre for Advanced Inter-
     net Architectures, Melbourne, Australia.  More details are available at:

     http://caia.swin.edu.au/urp/newtcp/

AUTHORS

     The cc_cubic congestion control module and this manual page were written by Lawrence Stewart <lstewart@FreeBSD.org> and David Hayes
     <david.hayes@ieee.org>.

BSD
								September 15, 2011							       BSD

CC_VEGAS(4)						   BSD Kernel Interfaces Manual 					       CC_VEGAS(4)

NAME

     cc_vegas -- Vegas Congestion Control Algorithm

DESCRIPTION

     The Vegas congestion control algorithm uses what the authors term the actual and expected transmission rates to determine whether there is
     congestion along the network path i.e.

	   actual rate = (total data sent in a RTT) / RTT

	   expected rate = cwnd / RTTmin

	   diff = expected - actual

     where RTT is the measured instantaneous round trip time and RTTmin is the smallest round trip time observed during the connection.

     The algorithm aims to keep diff between two parameters alpha and beta, such that:

	   alpha < diff < beta

     If diff > beta, congestion is inferred and cwnd is decremented by one packet (or the maximum TCP segment size).  If diff < alpha, then cwnd
     is incremented by one packet.  Alpha and beta govern the amount of buffering along the path.

     The implementation was done in a clean-room fashion, and is based on the paper referenced in the SEE ALSO section below.

IMPLEMENTATION NOTES

     The time from the transmission of a marked packet until the receipt of an acknowledgement for that packet is measured once per RTT.  This
     implementation does not implement Brakmo's and Peterson's original duplicate ACK policy since clock ticks in today's machines are not as
     coarse as they were (i.e. 500ms) when Vegas was originally designed.  Note that modern TCP recovery processes such as fast retransmit and
     SACK are enabled by default in the TCP stack.

MIB Variables
     The algorithm exposes the following tunable variables in the net.inet.tcp.cc.vegas branch of the sysctl(3) MIB:

     alpha  Query or set the Vegas alpha parameter as a number of buffers on the path.	When setting alpha, the value must satisfy: 0 < alpha <
	    beta.  Default is 1.

     beta   Query or set the Vegas beta parameter as a number of buffers on the path.  When setting beta, the value must satisfy: 0 < alpha <
	    beta.  Default is 3.

SEE ALSO

     cc_chd(4), cc_cubic(4), cc_hd(4), cc_htcp(4), cc_newreno(4), h_ertt(4), mod_cc(4), tcp(4), khelp(9), mod_cc(9)

     L. S. Brakmo and L. L. Peterson, "TCP Vegas: end to end congestion avoidance on a global internet", IEEE J. Sel. Areas Commun., 8, 13,
     1465-1480, October 1995.

ACKNOWLEDGEMENTS

     Development and testing of this software were made possible in part by grants from the FreeBSD Foundation and Cisco University Research Pro-
     gram Fund at Community Foundation Silicon Valley.

HISTORY

     The cc_vegas congestion control module first appeared in FreeBSD 9.0.

     The module was first released in 2010 by David Hayes whilst working on the NewTCP research project at Swinburne University of Technology's
     Centre for Advanced Internet Architectures, Melbourne, Australia.	More details are available at:

     http://caia.swin.edu.au/urp/newtcp/

AUTHORS

     The cc_vegas congestion control module and this manual page were written by David Hayes <david.hayes@ieee.org>.

BSD
								September 15, 2011							       BSD