Learn how to design and build parallel programs, including programs that employ multiple processors running at once to solve a large computational problem, with the clear presentation found in BUILDING PARALLEL PROGRAMS, International Edition. Written by experienced instructor and industry developer Alan Kaminsky, this book explores techniques for parallel programming on both major categories of parallel computers: SMPs and clusters. You gain valuable first-hand experience in working with the increasingly popular programming language, Java, as you complete programs from the text and work with a special Java class library, ideal for first-time parallel programmers. The author even teaches you how to effectively use performance metrics to design solid parallel programs, which is not addressed in most other books.
Perfect the contemporary skills and gain the understanding you need for success in parallel programming with Kaminsky's BUILDING PARALLEL PROGRAMS, International Edition today.
Part I Preliminaries
- Chapter 1 Parallel Computing
- Chapter 2 Parallel Computers
- Chapter 3 How to Write Parallel Programs
- Chapter 4 A First Parallel Program
Part II SMPs
- Chapter 5 Massively Parallel Problems
- Chapter 6 SMP Parallel Programming
- Chapter 7 Massively Parallel Problems, Part 2
- Chapter 8 Measuring Speedup
- Chapter 9 Cache Interference
- Chapter 10 Measuring Sizeup
etc.
Part III Clusters
- Chapter 19 A First Cluster Parallel Program
- Chapter 20 Parallel Message Passing
- Chapter 21 Massively Parallel Problems, Part 3
- Chapter 22 Data Slicing
- Chapter 23 Load Balancing, Part 2
- Chapter 24 Measuring Communication Overhead
- Chapter 25 Broadcast
- Chapter 26 Reduction, Part 3
- Chapter 27 All-Gather
- Chapter 28 calability and Pipelining
etc.