[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Maximum Advertised Window in TCP-Reno (Cont'd)
Thanks for the reply, but I still don't understand (or
agree). Following the ns documentation, window_ is the
maximum bound on the window size. I don't see two
windows in ns, or in TCP for that mattter.
The think I don't understand is how maxcwnd_ in tcp.cc
is set at run time. I see the command bind("maxcwnd_",
&maxcwnd_), but I don't know what how to specify a
value for this "maxcwnd_". I thought it is done by
setting window_, but apparently it doesn't work that
way.
Regards,
Hussein.
--- Peerapol Tinnakornsrisuphap <[email protected]>
wrote:
> Hi,
>
> I think window_ is the maximum 'flow control'
> window, but cwnd_ is the
> 'congestion' window. Congestion window will keep
> growing as long as the
> sending TCP keep getting ACK without lost. But the
> actual transmitted
> traffic will also be limited by maximum 'flow
> control' window (or receiver's
> advertised window).
>
> Peerapol.
>
> ----- Original Message -----
> From: hh hh <[email protected]>
> To: <[email protected]>
> Sent: Wednesday, November 10, 1999 10:37 PM
> Subject: Re: Maximum Advertised Window in TCP-Reno
> (Cont'd)
>
>
> >
> > Hi,
> >
> > After looking at the nam trace of my simulation,
> it
> > looks like the window size is behaving as
> expected.
> > (window increase stops at the specified window_).
> > However, when I look at the data recorded using a
> > "record" procedure similar to that in Marc Greis
> > tutorial, it shows that the cwnd_ increases
> without
> > bound!
> >
> > Would appreciate your comments,
> >
> > Hussein.
> >
>
>
>
=====
__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com