Hello...........


I will be more precise on my mathematical proof using probability...

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 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.

But since when the serial part of the Amdahl's law is bigger
you will hit probabilistic more contention, and since we can
make an approximation that smaller serial parts of Amdahl's law
are distributed probabilistically in a such way that you will
get more chance probabilistically that you will get farther from the
smaller serial parts of the Amdahl's law, so this prooves mathematically
that USL methodology is successful and is
an amazing great tool that can predict scalability.


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)