Usually a book’s title will give you a good idea what to expect. Unfortunately, the embedded and multi-core aspects of this title are somewhat general. It does cover some of the benefits of using multicore systems but the bulk of the coverage addresses Intel’s architecture and related optimization.
Still, if your expectations are to find out more parallel programming issues and solutions for Intel’s x86 architecture, then this book may be what you are looking for. It takes a look at tools like Intel’s Thread Building Block (TBB) and OpenMP—a multiplatform, shared-memory parallel programming system. These are the kind of tools that are targeting multi-core systems and embedded applications can make use of them but they work equally well for PCs running single core processors, assuming you can find one.
Dommeika starts out with a number of chapters that set the multicore stage. It does so at a relatively high level. The meat of the book covers scalar and parallel optimizations. Some case studies follow these chapters, where data decomposition and functional decomposition are explored. The open source AMIDE (A Medical Imaging Data Examiner) project is the subject at hand and how it takes advantage of OpenMP. The functional decomposition takes a look at SNORT—an open source intrusion detection system. These chapters do a good job of presentation tools like OpenMP in context.
I liked the chapter on Virtualization and Partitioning. The view is high level and it covers VirtualLogix’s VLX, an embedded virtual machine manager. Of all the chapters, this is the one that matches the title’s intent best.
The book wraps up with coverage of debugging and trends.
Overall, the content and presentation was top notch. It examines x86 parallel programming at a high level. So if this is what you are looking for then you have the right book.
