- We have seen a program to
find the minimum value
in an array before
( click here )
- We will
write the same program - but using the OpenMP pragmas - The structure of the program is as follows:
/* Shared Variables */ double x[1000000]; // Must be SHARED (accessed by worker threads !!) int start[100]; // Contain starting array index of each thread double min[100]; // Contain the minimum found by each thread int num_threads; int main(...) { for (i = 0; i < MAX; i++) x[i] = random()/(double)1147483648; #pragma omp parallel { // Note: we use omp_get_thread_num() to let a thread // figure out by itself what portion of work to process ... Thread i finds its minimum and ... store the result in min[i] } // ---------------------------------------- // Post processing: Find actual minimum // ---------------------------------------- my_min = min[0]; for (i = 1; i < num_threads; i++) if ( min[i] < my_min ) my_min = min[i]; }
- The code of the parallel region is as follows:
/* Shared Variables */ double x[1000000]; // Must be SHARED (accessed by worker threads !!) int start[100]; // Contain starting array index of each thread double min[100]; // Contain the minimum found by each thread int num_threads; int main(...) { for (i = 0; i < MAX; i++) x[i] = random()/(double)1147483648; #pragma omp parallel { int id; int i, n, start, stop; double my_min; n = MAX/omp_get_num_threads(); // step = MAX/number of threads. id = omp_get_thread_num(); // id is one of 0, 1, ..., (num_threads-1) start = id * n; if ( id != (num_threads-1) ) { stop = start + n; } else { stop = MAX; } my_min = x[start]; for (i = start+1; i < stop; i++ ) { if ( x[i] < my_min ) my_min = x[i]; } min[id] = my_min; // Store result in min[id] } // ---------------------------------------- // Post processing: Find actual minimum // ---------------------------------------- my_min = min[0]; for (i = 1; i < num_threads; i++) if ( min[i] < my_min ) my_min = min[i]; }- Example Program: (Demo above code)
- Prog file: click here
Compile with:
-
g++ -fopenmp openMP-min.C
Run with (on compute):
export OMP_NUM_THREADS=8
a.out
- Note:
- We can also
make the threads process the array elements
in different order
Example:
values handled by thread 0 | | | | | | | | | | | | | | V V V V V V V V V V V V V V |-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-| ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ | | | | | | | | | | | | | | values handled by thread 1 Thread 0 Thread 1 | | | | V V min[0] min[1] \ / \ / \ / \ / \ / main thread | | V Actual minimum
-
Each thread would execute
the following code:
my_min = x[start]; for (i = start+num_threads; i < MAX; i += num_threads ) { if ( x[i] < my_min ) my_min = x[i]; }
- Due to paging (accessing the large arrays by unsynchronized threads), the running time is poorer than the previous example
- Example Program: (Demo above code)
- Prog file: click here
How to run the program:
- Right click on link and
save in a scratch directory
- To compile: g++ -fopenmp openMP-min2.C
- To run: ./a.out
- The structure of the program is as follows: