[Info-vax] Storage is faster than the processor

[Jan-Erik Soderholm]

Den 2016-01-07 kl. 13:56, skrev johnson.eric at gmail.com:
> On Wednesday, January 6, 2016 at 5:38:24 PM UTC-5, Jan-Erik Soderholm
> wrote:
>
>> Shouldn't "figure 2" have "ms" (instead of ns) on the Y axes?
>
> Figure 2 is definitely fuzzy. Not clear what is meant by "packet". An
> MTU or a full sized IP packet? And what is "processing"?
>
> Regardless, I expected the time units to be microseconds (us), not
> milliseconds or nanoseconds.

Hm, a traditional disk has seek times in the 10s of milliseconds
area, not microseconds. The chart shows "10". And that also matches
the 10.000 ns value shown in chart 1.

Anyway, I find the whole article a bit like "kicking in open doors".

Jan-Erik.


> And even separate from that issue, I think
> most everyone here can easily understand that 10GigE will be faster than
> 1GigE by a factor of 10. So a chart isn't really needed for that.
>
> EJ
>