Wednesday, April 13, 2016

Digest for comp.programming.threads@googlegroups.com - 13 updates in 8 topics

roseycella217@gmail.com: Apr 13 12:55AM -0700

Volive Solutions ecommerce website designer in Saudi has been credited by previous clients, this gives us headstart beyond other custom ecommerce website developments. Volive designer normally hаvе ѕuffiсiеnt еxроѕurе to аll еlеmеntѕ of eCommerce website dеѕigning. Thе result that уоu should еxресt either ways is a successful wеbѕitе thаt has thе ability to drаw prospective сuѕtоmеrѕ аnd dеlivеr ѕignifiсаnt соnvеrѕiоnѕ.
 
http://www.volivesolutions.com/ecommerce-website-design-companies.html
Ramine <ramine@1.1>: Apr 12 06:34PM -0700

Hello,
 
As you have noticed i am a computer programmer, but i have
also a a diploma in micro-electronics and informatics..
what i have learned also during all those years is
that programming is also an interesting field, especially
when you add and complete it with Operational research courses
as i am doing now, so as you have noticed i am not doing
just programming but also operational research, and
as you have seen me doing i have implemented like an architect
my USL programs in front of your eyes, and i think that my USL
programs are also part of Operational research , because they
predict scalability and optimize the criterion of the Cost and
that's there strong advantages, other than that you have seen
me doing mathematical polynomial regression in my USL programs
and also approximating well the derivative of the USL equation
iteratively when both coefficient Beta and Alpha of USL are greater
than zero, and also approximating the derivative of USL
when the coefficient Beta equal zero and coefficient Alpha
is greater than zero, so finally i have finished this
project that is my USL programs that come with the source code
and that i think are stable and more robust now, also i have
included the 32 bit and 64 bit windows executables of my programs
so that to easy the job for you.
 
 
 
Thank you,
Amine Moulay Ramdane.
bleachbot <bleachbot@httrack.com>: Apr 12 08:49PM +0200

bleachbot <bleachbot@httrack.com>: Apr 12 09:37PM +0200

bleachbot <bleachbot@httrack.com>: Apr 12 09:38PM +0200

bleachbot <bleachbot@httrack.com>: Apr 12 11:03PM +0200

bleachbot <bleachbot@httrack.com>: Apr 13 12:32AM +0200

Ramine <ramine@1.1>: Apr 12 05:06PM -0700

Hello,
 
 
We have to be smart...
 
If you download and read the free book about USL from here:
 
https://www.vividcortex.com/resources/universal-scalability-law/
 
You will notice that for the database system he is calculating
the mean response-time, that means that as you will higher
the number of threads the response-time will lower..but his
way of doing is not the right way, because when he has calculated
the mean response-time that respect the specified SLO, he is
limiting the throughput , but this is wrong, because let's take
for example an Ecommerce website that uses a database system, we have
to maximize the throughput and limit and control the waiting-time
programatically in this Ecommerce website, this is not the same as he is
doing, it is why i have explained to you how to do solve this problem
and how to do it in my previous posts..
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Apr 12 03:39PM -0700

Hello.....
 
We have to be smart please..
 
If you take a look at the help file of the USL R package here:
 
https://cran.r-project.org/web/packages/usl/usl.pdf
 
And if you take a look at this R program from Dr Gunther
the author of USL:
 
http://www.perfdynamics.com/Classes/Materials/USLcalc.r
 
 
You will notice that they are using nonlinear regression
to predict scalability, but when you are using
nonlinear regression it can give different coefficients
from those of the polynomial regression that gives the right
coefficients, so this is a limitation in the nonlinear regression
method, so i have implemented my USL programs using polynomial
regression and this gives the right coefficients Alpha and Beta.
 
So now i think that my USL programs are stable and more robust.
 
You can download the new version 2.8 of my programs from:
 
https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Apr 12 03:38PM -0700

We have to be smart please..
 
If you take a look at the help file of the USL R package here:
 
https://cran.r-project.org/web/packages/usl/usl.pdf
 
And if you take a look at this R program from Dr Gunther
the author of USL:
 
http://www.perfdynamics.com/Classes/Materials/USLcalc.r
 
 
You will notice that they are using nonlinear regression
to predict scalability, but when you are using
nonlinear regression it can give different coefficients
from those of the polynomial regression that gives the right
coefficients, so this is a limitation in the nonlinear regression
method, so i have implemented my USL programs using polynomial
regression and this gives the right coefficients Alpha and Beta.
 
So now i think that my USL programs are stable and more robust.
 
You can download the new version 2.8 of my programs from:
 
https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Apr 12 02:51PM -0700

Hello,
 
Universal Scalability Law for Delphi and FreePascal was updated to
version 2.8
 
If you take a look at the following page of this R program that
implements USL:
 
http://www.perfdynamics.com/Classes/Materials/USLcalc.r
 
First, you have to know that my programs are different from this R program..
 
But you will notice that it prints the coefficients of USL
using the %.4f and %.6f formatting, but that's not
the right way to do it, so i have implemented
something more efficient, and here is what i have done in
Delphi and FreePascal, here is the right formatting of the
coefficients of USL:
 
--
if strtofloat(Format ('%.7f', [alpha])) > 0
then writeln('The coefficient Alpha of USL is: ', Format ('%.7f', [alpha]))
else if strtofloat(Format ('%.7f', [alpha])) < alpha
then writeln('The coefficient Alpha of USL is: ', alpha)
else if strtofloat(Format ('%.7f', [alpha])) = 0.0
then writeln('The coefficient Alpha of USL is: ', Format ('%.7f',[alpha]));
 
if strtofloat(Format ('%.7f', [beta])) > 0
then writeln('The coefficient Beta of USL is: ', Format ('%.7f',[beta]))
else if strtofloat(Format ('%.7f', [beta])) < beta
then writeln('The coefficient Beta of USL is: ', beta)
else if strtofloat(Format ('%.7f', [beta])) = 0.0
then writeln('The coefficient Beta of USL is: ', Format ('%.7f',[beta]));
--
 
 
Also i have changed real type to double type and i have changed
the peak2 variable to an int64 type.
 
So now i think that my USL programs are stable and more robust.
 
You can download version 2.8 of my programs from:
 
https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal
 
 
 
Thank you,
Amine Moulay Ramdane.
wij@totalbb.net.tw: Apr 12 09:57AM -0700

Description:
Xn project tries to develop a three-in-one 'language'. Xn is capable
of almost all kinds of data storage, communication, even executable
scripts. Applicable cases are therefore very general. Take a larger
imaginable one as example, an AI-network (even alive) can't be far
away. But all the important thing is that Xn is itself simple. Most
usecases, not so ambitious, various average programming tasks or
tools can use Xn to shorten their development time.
 
This short period of release time (3-weeks) has shown that the
implement (basic Xn parsing) is not a big job. The big thing is idea.
Xn is published in Fair Public Domain(convention). Any programming
language can implement it easily or tryout and share your better idea.
 
The purpose, except populate this project, is seeking for suggestions,
ideas, information or web-links to avoid err-and-try and make better
future decision. For instance, I don't know Qt very well, but feel Xn
can be made to produce Qt widget easily; And not really knowing any
real parallel computation, but feels it should be simple with Xn...
may be just try to find out the proper next step from you.
wij@totalbb.net.tw: Apr 12 10:00AM -0700

Xn project (version 0.04) is released.
https://sourceforge.net/projects/systemnode/
 
Description:
Xn project tries to develop a three-in-one 'language'. Xn is capable
of almost all kinds of data storage, communication, even executable
scripts. Applicable cases are therefore very general. Take a larger
imaginable one as example, an AI-network (even alive) can't be far
away. But all the important thing is that Xn is itself simple. Most
usecases, not so ambitious, various average programming tasks or
tools can use Xn to shorten their development time.
 
This short period of release time (3-weeks) has shown that the
implement (basic Xn parsing) is not a big job. The big thing is idea.
Xn is published in Fair Public Domain(convention). Any programming
language can implement it easily or tryout and share your better idea.
 
The purpose, except populate this project, is seeking for suggestions,
ideas, information or web-links to avoid err-and-try and make better
future decision. For instance, I don't know Qt very well, but feel Xn
can be made to produce Qt widget easily; And not really knowing any
real parallel computation, but feels it should be simple with Xn...
may be just try to find out the proper next step from you.
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.programming.threads+unsubscribe@googlegroups.com.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)