The book focuses on widely used APIs that enable parallel programming:
Point-to-point communication (send/receive), collective communication (broadcast, scatter, gather, reduce), and managing network latency.
The latter halves of the textbook walk readers through concrete implementations of parallel algorithms, such as: Parallel Computing Theory And Practice Michael J Quinn Pdf
Explores processor arrays, multiprocessors (shared memory), and multicomputers (distributed memory). Part 2: Practical Implementation Programming Languages (Ch 4): Surveys languages like Fortran 90, C*, Linda, and OCCAM. Mapping and Scheduling (Ch 5):
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. The book focuses on widely used APIs that
: Quinn surveys historically significant and popular architectures, including the Thinking Machines CM-5 and Intel Paragon , to illustrate how hardware design influences software choices. Key Chapters and Content
Combining tasks to improve performance and reduce overhead. Mapping and Scheduling (Ch 5): Quinn argues that
The PDF version of "Parallel Computing Theory and Practice" by Michael J. Quinn is a valuable resource for students, researchers, and practitioners in the field of parallel computing. The book provides a comprehensive introduction to the theory and practice of parallel computing, making it an ideal textbook for courses in parallel computing.
Managing communication between processors to prevent data race conditions. Parallel Algorithm Complexity
Managing race conditions, mutual exclusion, and cache coherence. Each processor has private memory; connected via a network. Message Passing Interface (MPI)
The theory of parallel computing provides a foundation for understanding the principles and limitations of parallel processing. It encompasses topics such as: