Hello..........


I am not crazy, please follow my proof...


I explain better my previous proof...

For a bigger part of the serial part of the Amdahl's law,
here is my proof:

If the serial part of the Amdahl's law is more bigger, so the
chance to get contention with the USL methodology with fewer
cores and fewer threads is bigger, so this will allow the USL
methodology to have a better approximation of the predicted
scalability and to forecast farther.

But i will be clear on the smaller parts of the Amdahl's law:

But if the serial part of the Amdahl's law is smaller , so the
chance to get contention with the USL methodology with fewer
cores and fewer threads will get smaller, so this will allow
the USL methodology to forecast farther with fewer cores and
fewer threads.

But since when the serial part of the Amdahl's law is bigger
you will hit probabilistically more the contention and this will
make the USL methodology work, and since the chance probabilistically
that the bigger parts that causes contention and the smaller serial
parts that are much farther from the serial part that causes contention
is much higher, so this this prooves mathematically that USL
methodology is successful and is an amazing great tool that can predict scalability and forecast scalability much farther.

This is my mathematical proof using probability.


So now you undertand that USL methodology is a really an amazing great
tool !

And my USL programs are a really amazing great tools !


You can download my USL programs version 3.0 with the source code from:

https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal


Thank you,
Amine Moulay Ramdane.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)