distributed computing implementations

[Steve Holden]

"Robin Becker" <robin at jessikat.fsnet.co.uk> wrote in message
news:6ZY$RKABo9l+Ew+N at jessikat.fsnet.co.uk...
> In article <b77hpo$59a$0 at 216.39.172.122>, Bengt Richter <bokr at oz.net>
> writes
> .....
> >>I really can't explain these numbers either? Localhost is 10 times
slower
> >>than over the network???
> >
> >IIRC, there was some discussion some time ago about this being a
particular
> >FreeBSD pessimality problem that wasn't true of Linux, but someone was
fixing
> >it?
> >But FreeBSD has good networking reputation I thought, so this needs
double
> >check,
> >for sure. Maybe it was for an old version?
> >
> ....this is version 4.5. I guess things would be easier if one could be
> sure that it's not a dns type thing, but I see from the dev list that
> there's already some discussion about how/where dns lookups can be
> avoided/speeded up with sockets which have a supposedly known address.
> Naively I thought it would be relatively easy to just get a 32 / 64 bit
> int back from some socket module function and then shove that into the
> address part of the socket address.
>
> Given the above perhaps it's not dns at all, but some other things that
> get in the way. Do OSes impose limits on number of ports or
> connections/second etc?
>

Seems that the easiest way to get insight into the network traffic you're
generating is to install Ethereal or some similar software and simply look
to see what's going over the wire.

regards
--
Steve Holden                                  http://www.holdenweb.com/
Python Web Programming                 http://pydish.holdenweb.com/pwp/
Did you miss PyCon DC 2003?            Would you come to PyCOn DC 2004?