> From [email protected] Thu Aug 7 13:35:18 1997
> From: "Farley, Tim" <[email protected]>
> To: "'Greg Miller'" <[email protected]>,
> "'TCP over Satellite'"
> <[email protected]>
> Subject: RE: Problematic Approaches
> Date: Thu, 7 Aug 1997 16:01:18 -0400
>
> >> Ah, but the good (?) news is that Eric's first bullet above comes
> into
> >> play and renders large windows useless for most Web traffic
> >> anyway. Our recent measurements show that the vast majority of Web
> >> connections are too short to benefit from window scaling and big
> >> socket buffers, regardless of the path.
>
> Sure, *now* that's the case. But should we really be basing our
> recommendations for future of TCP on the behavior of HTTP/1.0? HTTP 1.1
> exists and is being deployed widely as we speak. If it works as
> designed (and the experimental numbers which have been published are
> correct), it should radically change the above.
Not quite as much as planned. Although the connection is reused,
the window is not just be 'left open' in the meantime.
When the source of data goes idle, some versions of TCP
slam the window shut (slow start restart after idle).
The reason is to avoid a line-rate burst.
Which means that, although we're still inside one connection,
the issue of window startup is still moot in that case.
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