OpenMP CUDA

Makefile

+# Makefile 
+
+# Source file
+SRC = helloWorld.c
+
+# Compiler
+CC = gcc
+
+# Executable name
+EXE = ./out
+
+all: $(EXE) 
+
+$(EXE): $(SRC)
+	$(CC) -fopenmp -o $(EXE) $(SRC) 
+
+clean:
+	rm -f $(EXE)
+
+run:
+	$(EXE) 

README.md

+# OpenMP
+
+![OpenMP Fork-join Model](image1.png)
+
+# CUDA
+
+* [CoLAB](https://colab.research.google.com/drive/1UP2TJvtJfgiU8x9g0iB8ffE5fj8B4aPV#scrollTo=tcS8TCGLz9m0)
+* [CUDA Tutorial](https://cuda-tutorial.readthedocs.io/en/latest/tutorials/tutorial01/)
+* [Machine Learning on GPU](https://hsf-training.github.io/hsf-training-ml-gpu-webpage/aio/index.html)
\ No newline at end of file

add_Vec.cu

+#include <stdio.h>
+#include <stdlib.h>
+__global__ void add(int *a, int *b) {
+a[0]+= b[0];
+}
+
+int main() {
+int a, b;
+
+// host copies of variables a, b
+int *d_a, *d_b;
+
+// device copies of variables a, b
+int size = sizeof(int);
+
+// Allocate space for device copies of a, b
+cudaMalloc(&d_a, size);
+cudaMalloc(&d_b, size);
+
+// Setup input values
+a = 5;
+b = 100;
+
+// Copy inputs to device
+cudaMemcpy(d_a, &a, size, cudaMemcpyHostToDevice);
+cudaMemcpy(d_b, &b, size, cudaMemcpyHostToDevice);
+
+// Launch add() kernel on GPU
+add<<<1,1>>>(d_a, d_b);
+
+// Copy result back to host
+cudaError err = cudaMemcpy(&a, d_a, size, cudaMemcpyDeviceToHost);
+  if(err!=cudaSuccess) {
+      printf("CUDA error copying to Host: %s\n", cudaGetErrorString(err));
+  }
+printf("result is %d\n",a);
+
+// Cleanup
+cudaFree(d_a);
+cudaFree(d_b);
+
+return 0;
+}
\ No newline at end of file

helloWorld.c

+#include "omp.h"
+#include "stdlib.h"
+#include "stdio.h"
+
+
+//simple hello world parallel method
+void startParallelHeloworld()
+{
+    int x=0;
+#pragma omp parallel num_threads(8)
+    {
+        
+        int id = omp_get_thread_num();
+        //printf("Number of thread = %d\n", id);
+
+        if (id == 0)
+           printf("Hello From Master: => Threads Count is: %d \n", omp_get_num_threads());
+        else
+            printf("Hello from Thread %d\n", id);
+        for(int i=0;i<1000;i++){
+            x++;
+        }
+    }
+    //printf("Done x=%d",x);
+}
+
+//SPMD
+void forExample(){
+    int i, k, N=20;
+    double A[N], B[N], C[N];
+    #pragma omp parallel for
+        for (i=0; i<N; i++) {
+            int id = omp_get_thread_num();
+            A[i] = B[i] + k*C[i];
+            printf("\t %d Tid: %d\n",i,id);
+        }
+}
+
+
+//Main method
+int main(int count, char *arg[])
+{
+    int nthreads;
+    nthreads = omp_get_num_threads();
+    //printf("Main Number of threads = %d\n", nthreads);
+    startParallelHeloworld();
+    printf("level1\n");
+    forExample();
+    printf("level2\n");
+    //sleep(1);
+    return 0;
+}