Alen Peacock wrote:
On 1/27/07, David Barrett <[EMAIL PROTECTED]> wrote:
Is it actually any cheaper to send a 0-byte UDP packet than to send
one that
is sized at the MTU?
In terms of CPU and bus burnage, small UDP packets are going to be
WAAAAAY more expensive than sending/receiving large UDP packets. It
isn't too hard to confirm this through simple experiments.
Agreed, but this is the answer to a different question. I think that
David (who may correct me if I'm wrong), wants to know "if I'm sending
packets regularly, is there any advantage to making sure it is as small
as possible, or is there no real difference if I just fill it all the
way up?" vs. the other equally good question that you've answered which
is "if I need to move X megabytes of data over the net, is there an
advantage to using the entire MTU for each in order to minimize the
number of packets sent?".
The answer to the first is "yes, there's some advantage to trying to
keep the size of your packets down by sending data efficiently, but this
advangtage depends on the link technology your packets run over... for
some, like ethernet LANs, there's a lot of per-packet overhead that
dominates, for others like a PPP modem connection, every byte counts"
The answer to the second is "absolutely, given that some links have
large per-packet overhead and most routers and hosts have large fixed
overheads of CPU time for each packet that arrives, you should use as
much of the MTU as you can for each packet you send... while making sure
that A) you don't fragment and B) you know what head-of-line blocking
is, and how it matters to your application (if at all)"
Matthew Kaufman
[EMAIL PROTECTED]
_______________________________________________
p2p-hackers mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/p2p-hackers
