cc_htcp4
H-tcp congestion control algorithm
Overview
The H-TCP congestion control algorithm was designed to provide increased throughput in fast and long-distance networks. It attempts to maintain 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. The window increase algorithm operates like NewReno for the first second after a congestion event, and then switches to a high-speed mode based on a quadratic increase function.
The implementation was done in a clean-room fashion, and is based on the Internet Draft and other documents referenced in the SEE ALSO section below.
MIB Variables
The algorithm exposes the following tunable variables in the net.inet.tcp.cc.htcp branch of the sysctl(3) MIB:
- adaptive_backoff
-
Controls use of the adaptive backoff algorithm, which is designed to keep network queues non-empty during congestion recovery episodes. Default is 0 (disabled).
- rtt_scaling
-
Controls use of the RTT scaling algorithm, which is designed to make congestion window increase during congestion avoidance mode invariant with respect to RTT. Default is 0 (disabled).
Acknowledgements
Development and testing of this software were made possible in part by grants from the FreeBSD Foundation and Cisco University Research Program Fund at Community Foundation Silicon Valley.
Releases
36 releases since FreeBSD 8.3-RELEASE; the text changed 24 times.