cc_cdg(4) 맨 페이지 - 윈디하나의 솔라나라

개요

섹션
맨 페이지 이름
검색(S)

cc_cdg(4)

CAIA-Delay  Gradient  (CDG)  is a hybrid congestion control algo‐
rithm which reacts to both packet loss and inferred  queuing  de‐
lay.   It  attempts  to  operate as a delay-based algorithm where
possible, but utilises heuristics to detect loss-based TCP  cross
traffic  and will compete effectively as required.  CDG is there‐
fore incrementally deployable and suitable for use on shared net‐
works.  During delay-based operation, CDG uses  a  delay-gradient
based probabilistic backoff mechanism, and will also try to infer
non  congestion  related packet losses and avoid backing off when
they occur.  During loss-based operation, CDG essentially reverts
to behaviour.  CDG switches to loss-based operation when  it  de‐
tects that a configurable number of consecutive delay-based back‐
offs  have had no measurable effect.  It periodically attempts to
return to delay-based operation, but will keep switching back  to
loss-based operation as required.  The algorithm exposes the fol‐
lowing  variables in the branch of the MIB: Current algorithm/im‐
plementation version number.  Delay-based window decrease  factor
as  a  percentage  (on  delay-based backoff, w = w * beta_delay /
100).  Default is 70.  Loss-based window  decrease  factor  as  a
percentage (on loss-based backoff, w = w * beta_loss / 100).  De‐
fault is 50.  Scaling parameter for the probabilistic exponential
backoff.   Default is 2.  Number of samples used for moving aver‐
age smoothing (0 means no smoothing).  Default is 8.   Number  of
consecutive  delay-gradient  based congestion episodes which will
trigger loss-based CC compatibility.  Default is  5.   Number  of
consecutive  delay-gradient based congestion episodes to hold the
window backoff for loss-based CC compatibility.   Default  is  5.
If non-zero, this enables an experimental mode where CDG's window
increase  factor  (alpha) is increased by 1 MSS every RTTs during
congestion avoidance mode.  (Setting to 1 results in the most ag‐
gressive growth of the window increase  factor  over  time.   Use
higher values for slower growth.)  Default is 0.  Development and
testing  of  this  software  were made possible in part by grants
from the FreeBSD Foundation and  The  Cisco  University  Research
Program  Fund,  a corporate advised fund of Silicon Valley Commu‐
nity Foundation.  The congestion control module first appeared in
The module was first released in 2011 by David Hayes whilst work‐
ing on the NewTCP research project  at  Swinburne  University  of
Technology's  Centre  for  Advanced  Internet Architectures, Mel‐
bourne,   Australia.    More   details    are    available    at:
http://caia.swin.edu.au/urp/newtcp/ The congestion control module
was written by This manual page was written by and The underlying
algorithm  and  parameter values are still a work in progress and
may not be optimal for some network scenarios.






















맨 페이지 내용의 저작권은 맨 페이지 작성자에게 있습니다.
RSS ATOM XHTML 5 CSS3