 |
 |
|
|
 |
Week 1 — Fundamentals: speedup, models, PRAM. Week 2 — Parallel algorithm design: prefix, matrix ops, sorting. Week 3 — Programming practice: MPI/OpenMP basics, synchronization. Week 4 — Performance tuning, profiling, advanced topics and projects.
The core of Quinn’s methodology lies in the rigorous analysis of parallel algorithms. He emphasizes that parallel computing is not simply about running tasks simultaneously; it is about managing the trade-offs between and the overhead of communication. Quinn utilizes the PRAM (Parallel Random Access Machine) model to teach the theoretical limits of computation, while introducing students to the concepts of scalability and efficiency . By focusing on data dependencies and synchronization, the text provides a blueprint for decomposing complex problems into smaller, concurrent tasks. [1, 3, 5] Bridging Theory and Practice Parallel Computing Theory And Practice Michael J Quinn Pdf
Quinn argues that data-parallel algorithms are often superior to control-parallel ones because they scale as a function of the problem size rather than remaining a constant. Week 1 — Fundamentals: speedup, models, PRAM
For clusters, MPI is the lingua franca. Quinn excels here by explaining (Single Program, Multiple Data) models. He contrasts blocking sends ( MPI_Send ) vs. non-blocking sends ( MPI_Isend ), tying those functions back to the theory of latency hiding. Classic algorithms covered include: Week 4 — Performance tuning, profiling, advanced topics
: The "Practice" side of the book hits when Quinn introduces the obstacles— communication overhead synchronization costs
