• Skip to Main Content
  • Go to the Website's Home
  • Go to Website News
  • Go to About this site
  • Skip to Audience Navigation Menu
  • About
    • Welcome to the College
    • Facts and Figures
    • Message from the Dean
    • Visit Us
    • College Administration
    • News Center
  • Research
    • Research Home
    • Departments
    • Centers & Labs
    • Faculty Directory
    • Opportunities for Undergraduates
  • Admissions
    • Undergraduate Admissions
    • Undergraduate Recruiting
    • Graduate Admissions
  • Academics
    • Departments
    • Undergraduate Programs and Degrees
    • Graduate Programs and Degrees
    • Course Guide/Bulletin
    • Teaching
    • Support Services
  • Departments
    • Aerospace Engineering
    • Atmospheric, Oceanic and Space Sciences
    • Biomedical Engineering
    • Chemical Engineering
    • Civil and Environmental Engineering
    • Electrical Engineering and Computer Science
    • Industrial and Operations Engineering
    • Interdisciplinary Professional Programs
    • Materials Science and Engineering
    • Mechanical Engineering
    • Naval Architecture and Marine Engineering
    • Nuclear Engineering and Radiological Sciences
  • Support the College
    • Giving
    • Involve Yourself
    • Corporate Relationships
  •   

College of Engineering

Center for Advanced Computing

  • Welcome
  • About
  • Getting Started
    • Getting Started
    • Learning About HPC
    • CAC Training Courses
  • Resources
    • Resources
    • Systems
    • Software
    • Storage
    • Networking
    • Visualization
    • Teragrid
  • Academics
  • Contact Us
  • Frequently Asked Questions
Follow Us On Twitter

Contact Information

[Contact] Center for Advanced Computing
College of Engineering
cac-support@umich.edu Link

Center for Advanced Computing  /  Resources  /  Software  /  CUDA

CUDA

Compile CUDA Code
PBS Example

 

CUDA is the compute engine in NVIDIA graphics processing units or GPUs, that is accessible to software developers through industry standard programming languages. Programmers use C for CUDA using the nvcc compiler.

Compiling with CUDA

Load the CUDA Module

module load cuda

Use module show cuda to see variables the module set to make compiling simpler.

For details on CUDA as a programming tool see our Nyx CUDA page.

PBS Example

To use cuda jobs must request GPU's. CUDA will not run correctly without a GPU, and users must set which GPU they are using.

#PBS -N cuda-test
#PBS -l nodes=1,mem=1gb,walltime=00:15:00,software=graphics
#PBS -m abe
#PBS -V

cat $PBS_NODEFILE

 #run our app, which is assinged a GPU by the CUDA library

./a.out

CSS 2.1 Valid
University of Michigan
  • Home
  • Contact Us
  • Site Map
  • Site Feedback
  • Accessibility
  • Emergency Procedures
  • Jump to top of the page
© 2009 College of Engineering, University of Michigan