Time: W 18:00 - 20:45 (lecture)

Location: ENG 288

Description

Parallel programming overview; NIVIDIA GPU architecture and CUDA PTX; CUDA programming and tool chain; threads, atomics, and memory model; parallel patterns, performance and productivity; optimization techniques; application case studies. 

_{Course Goals and Student Learning Objectives}

Upon successful completion of this course, students will be able to:

1.     Acquire knowledge required by parallel programming – principles and patterns of parallel programming, GPU architecture features and constraints, programming API, tools, and techniques

2.     Obtain hands-on experience on how to program massively parallel processors to achieve high performance, as well as functionality, maintainability and scalability