[ih] Significant milestones in the history of TCP/IP

[dhc2 at dcrocker.net (Dave Crocker)]

On 9/17/2015 10:11 AM, Noel Chiappa wrote:
> I suspect the only way to say with any certainty how well a network built out
> of lots of slow lines, as opposed to a few fast ones, would have worked is a
> comprehensive simulation. Which is not likely to happen, of course!


First, lots of smaller lines means a higher percentage of the bandwidth
devoted to headers and control chatter.

Second, bigger pipes means shorter transit time, to the end of the
packet.  I vaguely recall having been told that this lower 'residency'
effect improved overall throughput, when comparing one big, to an
equivalent aggregate of multiple smaller.  I thought this was being
claimed as a statistical effect, but wasn't listening closely...

d/

-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net