The book contain 296 pages only, ( I doubt it is going to provide an in-depth knowledge from basics to advance).
Table of Contents
- Introduction to Parallel Programming
- Introduction to OpenCL
- OpenCL Device Architectures
- Basic OpenCL Examples
- Understanding OpenCL’s Concurrency and Execution Model
- Dissecting a CPU/GPU OpenCL Implementation
- OpenCL Case Study: Convolution
- OpenCL Case Study: Video Processing
- OpenCL Case Study: Histogram
- OpenCL Case Study: Mixed Particle Simulation
- OpenCL Extensions
- OpenCL Profiling and Debugging
Looking at the ToC it gives me some relief that the book has 5 case studies. I hope that the authors have put great efforts in explaining optimization for getting maximum acceleration (else for what we are doing OpenCl?).
From the ToC I can see that GPU architecture part is missing (or may be it is in side some of the chapters mentioned in the ToC). I thing for using a GPU for GPGPU, a good understanding of GPU architecture is a must.
Another thing that I am noticing is there is perhaps less talk on Graphics. A case study on Graphics will be useful.
Its good that there is a dedicated chapter on WebCL
How useful is the book will become clear once it is in the hands of the readers, but no doubt its filling the current void in the field of GPU Computing.
A similar book by Fixtars: “The OpenCL Programming Book” is already in the market and its HTML version is absolutely free.