Dive into the world of parallel computing through this comprehensive tutorial series spanning 4-5 hours. Explore key concepts starting with an introduction and performance overview, then delve into optimizing single-core performance through serial performance techniques and Single Instruction, Multiple Data (SIMD) operations. Progress to advanced parallel strategies, including multithreading, parallel algorithm design, tasks (co-routines), distributed computing, and GPU utilization. Gain a solid foundation in parallel computing principles and practical implementation strategies to enhance computational efficiency and performance.
Overview
Syllabus
Introduction
- Introduction (10:24)
- Performance Overview (27:16)
Optimizing Single-Core Performance
- Serial Performance (43:49)
- Single Instruction, Multiple Data (27:36)
Parallel Strategies
- Multithreading (52:04)
- Parallel Algorithm Design (14:53)
- Tasks (Co-routines) (19:23)
- Distributed Computing (66:23)
- GPUs (30:33)
Taught by
Dr. Matt Bauman
