PEANO order

[mia023]

hello everyone I have a project and just need ideas about it. This is the documentation:

[FONT=Calibri][SIZE=3][FONT=Calibri][SIZE=3]

Develop the code for matrix multiplication in Peano[/SIZE][/FONT][/SIZE][/FONT][FONT=Calibri][SIZE=3][FONT=Calibri][SIZE=3]‐[/SIZE][/FONT][/SIZE][/FONT][FONT=Calibri][SIZE=3][FONT=Calibri][SIZE=3]Order and another one in roworder.
The program should be able to read two matrices from a file, and output the
multiplication of these two matrices using both orders. Use PAPI (Performance
Application Programming Interface) to instrument your code. Your code should monitor
the following: L1 cache misses, L2 cache misses, number of FLOPS and total time.
Present these results in a table for different sizes of the matrices, analyze and justify

these results.
[/SIZE][/FONT][/SIZE][/FONT]

[donbock]

Are you using the Haskell language?

[mia023]

actually no i am using C language

[HabibBhutto]

Hi Dear,

just refer the book
C++ Projects
by Reeta Sahoo
publisher: khanna book publishing co. (p) Ltd

there is a project in this book with respect to matrix manipulation he/she demonstrated all basic operation with respect to matrix... !

It will help u alot... !

:) pray for me if my sugesstion works for u.. !

[mia023]

you should buy the book or it is free

[mia023]

it isn't working out need some hepl please

[mia023]

I want to write a C++ program regarding matrix multiplication using Peano order can you please help

[JosAH]

I merged your two threads: they're both about the same problem. A question for you: what have you done so far?

kind regards,

Jos (moderator)

[mia023]

this is my code but i think there is a problem in it


#include <iostream>
#include <string>
#include <vector>
#include <sstream>

using namespace std;
// NOTE: CHANGE THESE VARIABLES TO MEET THE REQUIREMENTS OF YOUR TESTCASE!!!!
#define numRowsA 3
#define numColumnsA 3
#define numRowsB numColumnsA
#define numColumnsB 3
static double toDouble ( string s);
static long toLong ( string s);
static int toInteger ( string s);

int MatrixA[9] = {1,2,3,4,5,6,7,8,9};
int MatrixB[9]= {10,11,12,13,14,15,16,17,18};
int MatrixMult[9];
int a = 0;
int b = 0;
int c = 0;
void peanomult(int phsA, int phsB, int phsC, int dim);
int main()
{
for (int i =0; i < 9; i++)
{
cout << MatrixA[i];
cout << " ";
}
cout << endl;
cout << endl;
for (int i =0; i < 9; i++)
{
cout << MatrixB[i];
cout << " ";
}
cout << endl;
cout << endl;

peanomult(0, 0, 0, 9);
for (int i =0; i < 9; i++)
{
cout << MatrixMult[i];
cout << " ";
}
cout << endl;
cout << endl;


return 1;
}


//void peanomult(int MatrixA, int MatrixB, int MatrixMult, int dim)
void peanomult(int phsA, int phsB, int phsC, int dim)
{
if (dim == 1)
{
MatrixMult[c] += MatrixA[a] * MatrixB[b];
}
else
{
peanomult( phsA, phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, -phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); a += phsA; b += phsB;
peanomult( phsA, phsB, -phsC, dim/3); a += phsA; c -= phsC;
peanomult( phsA, -phsB, -phsC, dim/3); a += phsA; c -= phsC;
peanomult( phsA, phsB, -phsC, dim/3); a += phsA; b += phsB;
peanomult( phsA, phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, -phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); b += phsB; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); a -= phsA; c += phsC;
peanomult( phsA, -phsB, phsC, dim/3); a -= phsA; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); a -= phsA; b += phsB;
peanomult( phsA, phsB, -phsC, dim/3); a -= phsA; c -= phsC;
peanomult( phsA, -phsB, -phsC, dim/3); a -= phsA; c -= phsC;
peanomult( phsA, phsB, -phsC, dim/3); a -= phsA; b += phsB;
peanomult( phsA, phsB, phsC, dim/3); a -= phsA; c += phsC;
peanomult( phsA, -phsB, phsC, dim/3); a -= phsA; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); b += phsB; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, -phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, phsB, phsC, dim/3); a += phsA; b += phsB;
peanomult( phsA, phsB, -phsC, dim/3); a += phsA; c -= phsC;
peanomult( phsA, -phsB, -phsC, dim/3); a += phsA; c -= phsC;
peanomult( phsA, phsB, -phsC, dim/3); a += phsA; b += phsB;
peanomult( phsA, phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, -phsB, phsC, dim/3); a += phsA; c += phsC;
peanomult( phsA, phsB, phsC, dim/3);
};
}

[mia023]

could anyone respond as soon as possible because the due date is real soon

[JosAH]

Wouldn't it be a good idea to at least supply a link about what that 'Peano order' is all about? Showing a piece of badly formatted code that doesn't work is not the way to go. Help us to help you.

kind regards,

Jos

[mia023]

http://www5.in.tum.de/~bader/publikat/matmult.pdf

[JosAH]

@mia023
So you simply copied that function from page #12 of that (interesting) article and you expect everything to automagically work? That is not much of a scientific attitude: read and understand that article; draw the trace of that curve on paper and step through that algorithm and see if the two follow the same trace.

kind regards,

Jos

[mia023]

I really don't know C++ well nor C the teacher asked us to do it and I read the article and found this code only so I figured it would be of help. I haven't taken Peano order before and therefore I didn't really understand the article well so please can you help me alittle bit at least.