Description
-
(10 pts.) Implement a “hello-world” program using OpenMP. Your program should create as many threads as their are processors/cores on your current machine. From each thread that you create, your program should print:
Hello from thread i.
This is <Your Name>’s first parallel program There are currently n threads running
Implement your program so that the text given above is printed in one block for each thread.
-
(10 pts.) Consider the following problem. Given an N×N matrix of numbers, an integer M, and a modulus Q, find the largest product of M modulo Q numbers in a row, either vertically, horizontally, or diagonally. Write a program that outputs the value of the largest product on a line by itself. For example, consider the following 5 × 5 matrix of values with the value M = 4 and Q = 1000000 :
10 1 1 1 1
-
10 11 12 13
-
1 101112
-
1
1
1
10
1
1
1
1
1
10
Your program should produce the answer “17160” (10 × 11 × 12 × 13). When M = 5, your program should answer “100000”
Implement your program in C++ and provide adequate documentation. Test in-put/output examples will be provided to help you debug your code.
-
(20 pts.) Use OpenMP to implement a parallel version of the program that you imple-mented in problem #2. Implement your program using #pragma omp parallel.
Compare your program’s running time to your original program. Your program should be at least 75% efficient on a four core machine on put of a 1000 x 1000 array with M = 500.
Implement your program in C++ and provide adequate documentation.
