Computer Architecture: Concepts and Evolution

by Gerrit A. Blaauw, Frederick P. Brooks Jr.

In this remarkable book on computer design, long-known in the field and widely used in manuscript form, Gerrit A. Blaauw and Frederick P. Brooks, Jr. provide a definitive guide and reference for practicing computer architects and for students. The book complements Brooks' recently updated classic, The Mythical Man-Month , focusing here on the design of hardware and there on software, here on the content of computer architecture and there on the process of architecture design. The book's focus on architecture issues complements Blaauw's early work on implementation techniques. Having experienced most of the computer age, the authors draw heavily on their first-hand knowledge, emphasizing timeless insights and observations. Blaauw and Brooks first develop a conceptual framework for understanding computer architecture. They then describe not only what present architectural practice is, but how it came to be so. A major theme is the early divergence and the later reconvergence of computer architectures. They examine both innovations that survived and became part of the standard computer, and the many ideas that were explored in real machines but did not survive. In describing the discards, they also address why these ideas did not make it. The authors' goals are to analyze and systematize familiar design alternatives, and to introduce you to unfamiliar ones. They illuminate their discussion with detailed executable descriptions of both early and more recent computers. The designer's most important study, they argue, is other people's designs. This book's computer zoo will give you a unique resource for precise information about 30 important machines. Armed with the factors pro and con on the various known solutions to design problems, you will be better able to determine the most fruitful architectural course for your own design.

Genres: Computer Science, Nonfiction, Science, Computers, Programming

1213 pages, Hardcover

First published March 2, 1997

Reviews

[Shawn Morel · Rating 5 out of 5]

This book should be part of the CS cannon.

This is the most amazing reference on computer architecture ever. Not quite as pedagogical in tone as Patterson and Hennessey (what seems to be the undergrad CS standard).

This book focuses much more on the why, the tradeoffs of computer architecture design (and to some extend implementation). It gradually helps build up an ontology and intuition of computer architecture in the reader's mind.

Hardware and software design is a game of tradeoffs. Exploring the history of attempted designs and hearing from someone who has first hand knowledge about what worked and didn't work, and importantly why a given design failed is crucial.

[William Blair · Rating 5 out of 5]

I can add nothing to the following comment, dated April 19th, 2010 at 7:06 PM, by someone who designated themselves "Anonymous." Hopefully, you will still be able to read the original submission at this web page. However, because good web sites with valuable information don't seem to hang around as long as they should, herewith is what I would have loved to have said:

I took Fred Brooks’ Advanced Computer Architecture course years ago in graduate school. My father said, quite correctly, that it was like taking a theology course from Jesus. This book, in its then-preprint form, was the text, and I eventually wore my photocopied version out completely. (Thank goodness it finally came out in conventional binding.)

I am now a software architect, doing internet work in Java. I don’t think I can work for an hour without referring to some concept that Fred taught in the course, or in this book.

Read it. Study it. The only thing that could be better would be to take the course from Fred.

I was privileged to have been a graduate student of Dr. Brooks; I worked for him as a research assistant, and took some courses taught by him. He was my thesis advisor. The anonymous comment could not have better encapsulated my feelings about his talents as a teacher, or this book. His other, long-famous book, The Mythical Man-Month: Essays on Software Engineering, 20th Anniversary Edition (TMMM), needs no review. His recent, soon-to-be-famous book, The Design of Design, will need no review (at least from me). This book stands alone as a towering intellectual accomplishment.

I still vividly remember -- long before TMMM was published and even before the System/360 edition of Automatic Data Processing was published -- a graduate student asking Dr. Brooks, "But how does a computer work?" What he meant was that it was not yet clear to him (and the rest of us could not admit it) just exactly how a computer stepped through instructions, fetching and executing them, and pushing bits and bytes around the 'data flow' in order to actually do that which the student asking the question already well understood it did. In other words, "How does the hardware execute instructions?" was really the question being asked. This book answers that long-ago question for many (important) computers, most of which did not yet exist when it was asked.

Then, computers famously filled large rooms and cost millions of dollars. Today, the processor in a cell phone has more computing power than all the computers in the world together had when I started programming. But I suspect that even more so today than then, most otherwise erudite computer geeks have no clue how the CPU or processor in a computer actually "works."

If you fit that description, and want to know the answer, it's in this book. But I warn you in advance, even just reading this book is like like reading a contemporary commentary on the Bible written by Jesus."

[Ashish Pradhan · Rating 5 out of 5]

good book

[Gopi Attada]

excellent