GPU high-performance computing-Cuda (China-pub)

GPU high-performance computing-Cuda (China-pub)

[Author] Zhang Shu; Yan yanli [same as the author's work]
[Release news agency] China Water Conservancy and hydropower press [book no.] 9787508465432
[Shelving time]
[Publication date] on December 16, October 2009 [Opening] [Page code] 276 [version times] 1-1
Sample chapter trial: http://www.china-pub.com/48582&ref=ps

Edit recommendations

Featured typical practical routines and detailed details on Cuda usage.
This article introduces the parallel programming method by combining theory with practice ..
In-depth analysis of the hardware architecture, prompting the ing between the model and the underlying layer
Carefully summarize the optimization experience and parse high-performance programming skills...

[Content Overview]
This book is the first book in China to fully introduce the Cuda software and hardware system architecture. This topic describes the syntax, hardware architecture, and program optimization skills required to use Cuda for general computing. It is an entry-level teaching material and reference book for general GPU computing program development ..

This book is divided into five chapters. Chapter 2 introduces the development history of General GPU computing, the history, current situation and problems of parallel computing, and Chapter 1st introduces the usage of Cuda, help readers understand the Cuda programming model, memory model, and execution model, and master the compiling methods of Cuda programs. Chapter 2 discusses the Cuda hardware architecture, in-depth analysis of the interaction between Tesla GPU architecture and Cuda general-purpose computing; Chapter 1 summarizes the advanced optimization methods of cuda, and explores tasks, memory access, and instruction flow efficiency; chapter 2 demonstrates how to use the powerful performance of Cuda to solve practical problems with rich instances...

This book can be used as an introduction to Cuda and a reference for programming. It is mainly intended for programmers and engineers engaged in high-performance computing and scientific researchers who use GPU to accelerate computing in specialized fields, and programmers interested in GPU general-purpose computing. Colleges and research institutions that offer relevant courses can also use this book as teaching materials...

 

[Directory information]

Preface.
Chapter 1 GPU general computing 1
1.1 multi-core computing development 2
1.1.1 CPU multi-core parallel 3
1.1.2 supercomputer, cluster, and distributed computing 4
1.1.3 CPU + GPU heterogeneous parallel 5
1.2 GPU development overview 8
1.2.1 GPU rendering assembly line 8
1.2.2 Shader Model 10
1.2.3 nvidia gpu development 11
1.3 from gpgpu to Cuda 12
1.3.1 traditional gpgpu development 12
1.3.2 Cuda development 13
Chapter 14 Cuda Basics 14
2.1 Cuda programming model 14
2.1.1 host and DEVICE 14
2.1.2 definition and call of the kernel function 15
2.1.3 thread structure 16
<View the detailed directory.