ca817be7
From: rick.jones2@hp.com
reuven wrote:
> It is well known that UDP packets may arrive out of order. It is my
> understanding that this occurs because each packet may take a
> different route.
> In reality, if I transmit UDP between two computers on the Internet
> (let's say they are in the same country), approximately how often
> will it happen that I get packet 2 before packet 1?
> My gut feeling is that this would occur very rarely. Maybe less than
> once an hour? Obviously, I am not looking for an exact answer, just
> an order of magnitude.
Frankly, that question is no different than:
It is well known that string can be any length. In reality how
long is a length of string?
If you have something that depends on packet order and you want to use
UDP, you have to ass-u-me that reordering is prevalent and so you have
to have a way to deal with the reordering. You might though be able
to ass-u-me that it is sufficiently rare that if the overhead of
dealing with a reordering is high it doesn't matter much.
Having said that... and feeling like handing people lengths
(unspecified) of rope :)
If you want to get some idea, you might try to hack some source to
emit a pair of ICMP Echo Requests back-to-back and see how often the
ICMP Echo Responses arrive in a different order. That presumes though
that there are enough systems "out there" which will respond to an
ICMP Echo Request. If you aren't already familiar with an ICMP Echo
Request then ignore the whole idea.
rick jones
--
oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...
--- SoupGate-Win32 v1.05
* Origin: you cannot sedate... all the things you hate (1:229/2)
|
