> On Jul 30, 2021, at 5:42 AM, Ian Swett <[email protected]>
> wrote:
>
> To frame this in statistical terms, the sum of multiple normal distributions
> is a normal distribution with the sum of the means and a sum of the
> variances(https://en.wikipedia.org/wiki/Sum_of_normally_distributed_random_variables).
> Since the standard deviation is the square root of the variance, the
> standard deviation is dominated by the largest source of variation, and if
> each link introduces some variance, it's really only the one that introduces
> the most variation that matters.
Sure, but that assumes independent distributions which does not much apply in
this case. But the point is still valid, there will likely be 1-2 problematic
hops in the entire ~10-20 e2e path.
> Given no link knows whether it's the dominant source of variation on the
> path, I'd suggest every link pass along packets as soon as possible.
Yes.
> Obviously, link-layer error correction and other approaches which increase
> reliability without introducing delay are still welcome and valuable.
+1, though improved reliability almost always comes at the cost of increased
delay (rtx, fec, etc.).
