Day 4

The focus of Day 4 is cuBLAS Vector Addition

Code Descriptions

cublas_vec_add.cu (Vector Addition using cuBLAS)

• Implements vector addition using the cuBLAS SAXPY operation.
• Avoids explicit cudaMemcpy by utilizing cublasSetVector and cublasGetVector.
• Uses the cuBLAS API for optimized GPU-accelerated BLAS operations.
• Performs C = A + B efficiently with cublasSaxpy.

Profiling and Running

To compile and run the CUDA program, use:

nvcc -o cublas_vec_add cublas_vec_add.cu -lcublas
./cublas_vec_add