> From [email protected] Thu Aug 7 18:18:38 1997
> Date: Thu, 7 Aug 1997 18:18:04 -0700 (PDT)
> From: Kacheong Poon <[email protected]>
> Subject: RE: Problematic Approaches
> To: [email protected]
> Cc: [email protected], [email protected]
>
> > idle = (tp->snd_max == tp->snd_una);
> > if (idle && tp->t_idle >= tp->t_rxtcur)
> > /*
> > * We have been idle for "a while" and no acks are
> > * expected to clock out any data we send --
> > * slow start to get ack "clock" running again.
> > */
> > tp->snd_cwnd = tp->t_maxseg;
> >
> >
> > i.e., if the unacknowledged data and max window sent are the same,
> > in this case, then the system IS idle (no timer!).
>
> tp->t_idle is reset to 0 when a segment is received. It is incremented by
> 1 by the slow timer. So only when new data is sent and t_idle is greater than
> or equal to the RTO, t_rxtcur, TCP restarts slow start. Am I missing something
So you have to keep sending it data in order for it not to restart?
PS - there's debate on whether this is the correct implementation of the
intent of the slow-start restart anyway. The idea is that if data isn't
ackd for a while, you _can_ burst into the net at the line rate, and
you need to avoid that.
Unless the source has unlimited data to send, there's going to be
a gap.
When the bw*delay is larger than the file, either:
the window wasn't open and it doesn't matter
the window was open and it slams
The window _does_ slam inbetween requests, regardless of how
fast you parse the incoming file, when the window is open large
and the bw*delay is large compared to the file.
At best, the window slams inbetween each user request. Which
is still a loss.
What am I missing?
Joe
----------------------------------------------------------------------
Joe Touch - [email protected] http://www.isi.edu/~touch/
ISI / Project Leader, ATOMIC-2, LSAM http://www.isi.edu/atomic2/
USC / Research Assistant Prof. http://www.isi.edu/lsam/
This archive was generated by hypermail 2b29 : Mon Feb 14 2000 - 16:14:29 EST