Prof. Alan Kaminsky
Rochester Institute of Technology — Department of Computer Science
Source Code Listings
Copyright © 2019 by Alan Kaminsky.
In the twenty-first century, scientists and engineers are tackling the world's
toughest computational problems with parallel computing. Using multiple
processor cores running simultaneously, parallel computers are solving these
problems in less time and with greater accuracy than ever before. Even desktop
PCs nowadays are powerful parallel computers. To take full advantage of the
capabilities of these machines, programmers must learn to write parallel
BIG CPU, BIG DATA teaches you how to write parallel programs for multicore machines, compute clusters, GPU accelerators, and big data map-reduce jobs, in the popular Java language, with the free, easy-to-use, object-oriented Parallel Java 2 Library. The book also covers how to measure the performance of parallel programs and how to design the programs to run as fast as possible.
With 42 years of industrial and academic computing experience, Alan Kaminsky has been teaching parallel computing since 2004. He is a professor in the Department of Computer Science at the Rochester Institute of Technology.
Read the Preface and Table of Contents.
Order the book from Barnes & Noble.
(Note: A prepublication draft of the book dated July 2018 was free, Creative Commons licensed. The published book is not free and is not licensed under the Creative Commons license or any other license.)
|Part II. Tightly Coupled Multicore||
Part III. Loosely Coupled Cluster
Chapter 18. Massively Parallel
Chapter 19. Hybrid Parallel
Chapter 21. Cluster Parallel Loops
Chapter 22. Cluster Parallel Reduction
Chapter 23. Cluster Load Balancing
Chapter 24. File Output on a Cluster
Chapter 25. Interacting Tasks
Chapter 26. Cluster Heuristic Search
Part IV. GPU Acceleration
Chapter 33. Objects on the GPU
– Class ZombieGpu2
– ZombieGpu2.cu kernel
Part V. Big Data
Chapter 37. Big Data Analysis
An archive is available containing all the illustrations in the book in PNG format.