Surviving the Design of Microprocessor and Multimicroprocessor Systems: Lessons Learned Review

Average Reviews:

(More customer reviews)I am extremely disappointed with this book. The coverage of the material lacks depth, thoroughness', and clarity. I had a bad feeling about this book from the point I read in the preface that the book uses "modern presentation methodologies", and "students claim that this book is fast to read and easy to comprehend".
This book is a fast read because it does not cover any material in depth. The majority of the book is devoted to tables and figures with captions that are in many cases half devoted to defining the acronyms used in the figure, or verbose repetition of the book's reading.
This book is further hindered by its attempt to be prophetic, "one billion transistors on a single chip!" The review on the back of the book claims "a lively personal account, complete with survival tips, of his experiences". There is nothing lively about this book. Personal experience is limited to references to Milutinovic's research papers, and the case study style material comprising the appendices of the book. Tips do not exist.
There is not enough depth, or comprehensive exploration of the material for "tips". There is nothing in the book that will even lead to insightful extrapolations by the reader. The depth of the book only leads the reader to obvious conclusions that may be covered later in the book, again without any depth.
The first six chapters are more or less half hearted explorations of the material they are intended to cover. It is not until the last two chapters, which are most closely and directly related to the authors premise for DSM (distributed shared memory) architectures, that the depth approaches any respectably level. Further, of the eight chapters that comprise the book, taking up 200 pages, the last chapter takes up 50 of those pages. This book would have to at least quadruple in page count before the material is covered in more than passing interest.
This book more or less reads like the authors cliff notes that he should have used to write the book. All "Advanced" sections are nothing more than runs of paragraphs containing superficial summaries of other research papers. The material is not timeless. It defeats the point of a having a book if you are better off reading the research papers it is derived from.
I found it even more annoying that the Milutinovic continually refers to himself in the third person throughout the book, even in the preface.
The only good thing about this book is the quality of the paper and binding. This is still detracted from by the lack of margin space in which to take notes.
This book is not worth the $10 it can be had for from a second hand source, much less the suggested price of $120 new.
If you do read this book, I recommend reading "UNIX Systems for Modern Architectures" by Curt Schimmel, and "the Cache Memory book" by Jim Handy in place of the first six chapters of this book. At that point you might as well find another book covering DSM systems and forget this one.

Click Here to see more reviews about: Surviving the Design of Microprocessor and Multimicroprocessor Systems: Lessons Learned

"This book … uniquely synthesizes Professor Milutinovi???’s thinking on the important issues in computer architecture … The result is a necessarily somewhat eclectic, personal statement by one of the leaders of the field." —Michael J. Flynn, Stanford University From the Foreword "How do we invest one billion transistors on a single chip?" asks Veljko Milutinovi??? as he ponders the ultimate goal of an entire distributed shared memory (DSM)—plus numerous specialized accelerators—on a single chip. He then goes on to present a lively personal account, complete with survival tips, of his experiences in the front line of the rapidly evolving arena of microprocessor and multimicroprocessor system design. Focusing on areas critical to the future of system-on-a-chip design, Milutinovi??? combines his unique perspective with authoritative discussions of cache, instruction level parallelism, prediction strategies, the I/O bottleneck, multithreading, and multiprocessors. He reinforces concepts using three case studies of his own computer system/accelerator implementations with additional details available through Web-based appendices. A key DSM concept, Reflective Memory System (RMS), and tools for evaluating new architectural ideas or characterizing applications are also covered in appendices. Designed for fast, easy comprehension, Surviving the Design of Microprocessor and Multimicroprocessor Systems integrates clear, up-to-date explanations with a wealth of figures and a thorough review of the technical literature. It brings readers up to speed with concepts essential for implementing their own ideas, while addressing the most important issues facing scientists and engineers in advanced computer design.

Buy cheap Surviving the Design of Microprocessor and Multimicroprocessor Systems: Lessons Learned now.