Introduction to GPGPU Computing and CUDA


May
16
2014

Add to Cal
  • Speaker: Seth Meyer
  • Host Department: Information Technology (LSAIT)
  • Date: 05/16/2014
  • Time: 10:00 AM - 11:30 AM

  • Location: 2001 LSA

  • Description:

    Please join us on Friday May 16th from 10:00–11:30am in room 2001 LSA. Seth Meyer from LSA will present on GPGPU computing and CUDA. Learn how and when your users can use their graphics cards to speed up their computations by up to a factor of 1000. In this seminar, we will provide an overview of CUDA (NVIDIA's platform for GPGPU computing), CUDA programming concepts, and several example applications. Additional topics will include sample code in C, Matlab, and R, how you can use CUDA at home, a survey of GPGPU resources available to U-M people, and a comparison of CUDA to other GPGPU technologies and Intel's Xeon Phi.

    As we've had relatively low attendance for research computing related topics in the past, I'll add: Even if you don't plan on doing anything with GPGPU computing yourself, please consider attending so you can learn what is and is not possible with GPGPU computing and when you might want to recommend it to your users.

    Note that the presentation will start a minute or two after 10:00 am, not on "Michigan time."

    This presentation is one in a series of monthly IT Community trainings on the third Friday of each month, see https://knowledgebase.umich.edu/Training. The trainings are open to the public, so feel free to forward this message to anyone who may be interested in attending.

    If anyone would like to volunteer to present a training topic to the community, just let us know and LSA can offer you our training room, projector, and Friday time slot. Or you can suggest and vote on ideas at https://www.google.com/moderator/#16/e=20521f.

    We're seeking presenters for June 20th, July 18th, and August 15th.