All,

Please consider having your undergraduate and graduate research students
participate in the following free class.  The class is offered 1 night a
week - the meeting day/time will be determined after students have
registered to accommodate as many students as possible.

*GPU Computing: Programming and Architecture*

Course provides an introduction to the architecture and programmability of
modern graphics processing units (GPUs), including the architecture used in
latest AMD and NVIDIA GPUs. Introduction to concepts of parallel
programming on GPUs with the use of OpenCL API in C/C++ environment. In
depth discussion of OpenCL host programming API and GPU code development.
Course work starts with development of basic programs in OpenCL, and
gradually evolves to complex code development. Discussion and exercises on
optimization techniques and use of vendor profilers for improving GPU code
performance. Class will include discussions on alternate GPU programming
languages such as CUDA, and OpenACC. Assignments will include code
development using OpenCL. Students will be provided accounts on
Northeastern GPU clusters to compile and execute code.

*Prerequisites:* C/C++ coding (must), Linux basics for compilation and
execution (optional, but will be a plus), parallel programming with
pthreads (optional, but will be a plus).

*Instructor:* Yash Ukidave (yukidave@ece.neu.edu)

*To register, send an email to Yash Ukidave.*
*________________________________________*
*Prof. David Kaeli*
*Dept. of Electrical and Computer Engineering*
*Northeastern University*
*333 Dana Research Center*
*Boston, MA 02115*
*(617)-373-5413*
*email: kaeli@ece.neu.edu <kaeli@ece.neu.edu>*