On Apr 11, 2011, at 2:13 PM, Andrew McNabb wrote:

> Certainly there is at least _some_ correlation between distance and
> packet latency.  After all, there is no way [that I am aware of] that a
> packet traveling 1000 miles could arrive faster than the time it takes light 
> to
> travel 1000 miles.

While what you say is likely true, I was proposing that there is no 
*appreciable* correlation between distance and packet latency, when there are 
multiple routers involved, because the router induced latency is so much 
greater than the latency induced by wire propagation delay.

I based this assertion on the fact that a router will add a *lot* more latency 
than a run of copper wire, even a very long run of switched copper wire[1]. 
Although I'm convincing myself as I write this now that a slow, switched 
network could add a lot of latency on its own too...

Given the fact that this was the mid-90s, it seems a bit more plausible. Back 
then, big nodes like universities had a lot fewer hops between them.

Ah, those were the good old days, before the spammers. And MySpace. If only I 
weren't so dumb back then -- I would have recorded latency measurements between 
my publicly routable IP address in my *dorm room* and other universities.

--Dave

[1] I'm waving my hands and not offering numbers. Suck it up.
--------------------
BYU Unix Users Group 
http://uug.byu.edu/ 

The opinions expressed in this message are the responsibility of their
author.  They are not endorsed by BYU, the BYU CS Department or BYU-UUG. 
___________________________________________________________________
List Info (unsubscribe here): http://uug.byu.edu/mailman/listinfo/uug-list

Reply via email to