Write a MPI - CUDA program, for computing the matrix -vector To write a Pthreads - Cuda program to compute the matrix-vector multiplication using block striped partitioning of a matrix for uniform data distribution.

This is an implementation of Matrix-Vector multiplication using the block striped partitioning algorithm. Each thread multiplies the corresponding elements of the matrix with the vector by calling the CUDA kernel MatrixVectorMultiplication and writes the result into the result vector. A Mutex is used on the result vector to guarantee atomicity. The thread accesses the elements based on its id which is allocated by the main thread in the order of their creation. As the number of threads and the number of elements is known, the corresponding elements to be accessed can easily be computed.

Step 1 : Three Arrays MatrixA , VectorB and ResultVector are declaed on . Two additional vectors are used on host-CPU for computation. The purpose is to to store the part of the matrix that is accessible by each thread and the other is to store part of the result vector computed by each thread.

Step 2 : Main thread initializes the arrays MatrixA , VectorB and ResultVector . Fill arrays MatrixA and VectorB with single precsion real values and initialize the array ResultVector.

Step 3 : ThreadPart which decides how many elements each thread is to access in the matrix is computed.

Step 4 : Threads creation is done.

Step 5 : Memory is allocated for MyMatrixA and MyResultVector by all the threads on host-CPU .

Step 6 : MyMatrixA is constructed from MatrixA depending on the value of thread Id host-CPU

Step 7 : Memor for Vectors are allocated on host-CPU . The values of the vectors in the host machine are copied to the vectors on the device machine.

Step 8 :Checking for the availability of device is done on host-CPU

Step 9 : Each thread computes the Matrix vector multiplication by calling the CUDA kernel MatrixVectorMultiplication and copies back the result to MyResultVector on host-CPU .

Step 10 : Each thread performs a mutex operation on the result vector and assigns the corresponding elements on host-CPU.

Step 11 : Main thread waits for all threads to exit and Main thread prints the resultant vector.

CUDA API used :

To Allocate memory on device-GPU :
cudaMalloc(void** array, int size)

To Free memory allocated on device-GPU:
cudaFree(void* array)

To transfer from host-CPU to device-GPU:
cudaMemcpy((void*)device_array, (void*)host_array, size, cudaMemcpyHostToDevice)

To transfer from device-GPU to host-GPU:
cudaMemcpy((void*)host_array, (void*)device_array, size, cudaMemcpyDeviceToHost)