I correct, read again....

[Ramine]

Hello....


Although i have given an intuitive proof about USL in my previous post...

I don't think that this intuitive proof is 100% safe..

So the best thing is to not forecast too far with my USL programs..

You can forecast scalability to about 3X or 4X the number of cores
of the maximum number of cores of the empirical performance data,
and this is also useful.

I think that's the best thing the USL methodology can do ,
and that's the best thing my USL programs can do.



Thank you,
Amine Moulay Ramdane.

[Ramine]

Hello....

About my previous post about mixed workload and eCommerce websites..

You have seen my previous general solution about this case..

I will make it more precise: if you want to apply the
USL methodology with my USL programs to mixed workload
of eCommerce websites, i think here is necessary
conditions:

1- the mean time of the inter-arrivals of the internet users is
assumed to be a good approximation.

2- the webserver database systems must be set with the
right number of cores and the right number of threads
that ensure a more linear scalability.

3- the internet network is assumed to have a more linear
scalability even if its derivative of its linear scalability is
negative.

So those necessary conditions permit the nonlinear regression
of my USL programs to predict scalability of eCommerce websites.

So that makes my USL programs an amazing great tools to foerecast
scalability, and it makes the USL methodology an amazing great tool.

[Ramine]

Hello......

USL methodology is amazing and it works !

My previous proof was not correct...

I will explain why USL methodology is amazing and it works !

The Amdahl's law is a rational function...

and when in a parallel program you have a locked region using for
example locks, when you will draw the scalability graph of this parallel
program , you will notice that it looks like the Amdahl's law rational
function with a corresponding serial part, this is why USL methodology
is working perfectly
and it predicts scalability and it doesn't have a limitation,because the
simulation of the empirical world in parallel programs look like the
rational function of the Amdahl's law.

So now you have finally understood that USL methodology is an amazing
tool that predict effectively scalability using nonlinear regression or
polynomial regression.

I have just updated my USL programs that works great to version 3.1...

I have included the 32 bit and 64 bit windows executables of my programs
inside the zip file to easy the job for you.

You can download my USL programs version 3.1 with the source code from:

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