SGeMM: NVIDIA's Most Important Function
Matrix Multiplication is probably the algorithm of the 21st Century.
GPU Accelerated Matrix Multiplication
Accelerated matrix multiplication using CUDA C/C++. To make things interesting, let us try to match the performance of NVIDIA cuBLAS.
What is SGeMM
SGeMM stands for Single-Precision General Matrix Multiplication. Let's analyze matrix multiplication on a CPU and a GPU.
Step 1: Getting Started with CUDA Programming
Parallel matrix multiplication using CUDA C++.
Step 2: GPU Global Memory Coalescing
Memory coalescing is the most crucial concept in GPU programming. With matrix multiplication, we can get upwards of 7x improvement.
Step 3: GPU Shared Memory
Tiled matrix multiplication using GPU shared memory.