> From: Eric Travis <[email protected]>
> Subject: Problematic Approaches
>
>
> Look forward to the day (hopefully real soon) when window-scaling
> is ubiquitous; As we remember, window-scaling does *not* mean
> large-windows (it allows it to happen, but the buffer space needs
> to be allocated on BOTH sides of a connection). Now, imagine:
>
> o You want to do something on the Internet that will benefit
> from large windows (say: transfer size > bandwidth*delay product)
A system with a 2 meg transfer can benefit from a 2 meg window,
even if the BW*delay is 100 meg. It might be better to state:
windowsize is 'large', even though:
windowsize <= min(transfersize, bw*delay product, buffers)
> In addition, if we try to run with windows that are greater than the
> bandwidth-delay product, we'll be continually knocking on congestion's
This never helps. On a lossless link, the window never needs to
be larger than the bw*delay product. On a lossy link, the window needs
to be _smaller_ than whatever it was when it was lossy, i.e., smaller
than that.
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