Any clues to source of this delay?

Markus Stenberg mstenber at
Tue Aug 3 23:38:59 EDT 1999

Skip Montanaro <skip at> writes:
>     Markus> Problem:
>     Markus> On Linux, single connections have 'glass ceiling' of 50
>     Markus> roundtrips/second; no matter what I kludge, it seems to stay
>     Markus> there. Only way to fix this was to send 'empty' packets after
>     Markus> processed ones, thus increasing throughput by about 5X.
>     Markus> On NT, the glass ceiling is same, just 1/10 of Linux's (5
>     Markus> roundtrips/second).
> TCP's slow start algorithm, perhaps?  If you can figure a way to reuse the
> connection for multiple messages you may find that subsequent messages are
> sent more quickly.  A quick check at Google suggests that slow start details 
> might be found in RFC's 2001 and 2581.

Oh, sorry, wasn't being explicit; it isn't slow start, as the ceiling is on
_single_ connection, however long it stays up. By opening multiple
connections, the server's total handling capacity goes up but single
connections' speed stays constant.


> Skip Montanaro	|
> skip at  |
> 847-475-3758

	Markus Stenberg

More information about the Python-list mailing list