Shodor

a national resource for computational science education

HOME UPEP Shodor Blue Waters

Matrix Multiplication with CUDA

By Rob Hochberg
Shodor, Durham, North Carolina

This module teaches:

  • Matrix multiplication in the context of enumerating paths in a graph
  • The basics of programming in CUDA
  • Emphasizes the power of using shared memory when programming on GPGPU architectures

Upon completion of this module students should be able to:

  • Write a simple program making use of CUDA
  • Appreciate the power of using shared memory
  • Multiply matrices and know why it's worth it

The documents can be downloaded individually, or as a zip archive containing all of the documents:

Resources:

Module Document : The module document in PDF format.

Matrix Multiplication Code : A zip file containing the code accompanying this module.

Matrix Multiplication Module Assessment Document : The Matrix Multiplication Module Assessment Document in PDF format.

NVIDIA CUDA C Programming Guide : The NVIDIA CUDA C Programming Guide posted with special permission from the NVIDIA corporation.

NVIDIA CUDA C Best Practices Guide : The NVIDIA CUDA C Best Practices Guide posted with special permission from the NVIDIA corporation.

The CUDA Compiler Drive NVCC : The CUDA Compiler Drive NVCC Guide posted with special permission from the NVIDIA corporation.