Using Multiple Cores and GPUs in Native Code

Ever wonder how to get full use of all those cores in your machines? How about using your GPU for more than just painting the screen? This talk will cover the various libraries that are available to make parallel programming in C++ easier. We will explore what is available in PPL, C++ AMP, and CUDA GPU. Most of the talk will focus on how to get the most out of GPUs using the CUDA libraries.

Speaker Details

Adam Eversole started working at Microsoft in 1989, has worked on various applications in the Office suite. (Publisher, Word, Excel, Access). He worked in the Japanese subsidiary for a few years, was a Consultant with Microsoft Consulting for a few years, and has been working for the last decade in the Advanced Development Team in MSR. For the last year he has been working on GPU-based machine learning projects.

Date:
Speakers:
Adam Eversole
Affiliation:
MSR

    • Portrait of Adam Eversole

      Adam Eversole

      Software Architect

    • Portrait of Jeff Running

      Jeff Running