Customer Reviews

Computer Architecture: A Quantitative Approach, 4th Edition

5 star:		(13)
4 star:		(9)
3 star:		(4)
2 star:		(3)
1 star:		(1)

 

 

Computer architecture has seen vast changes in the last 20 years, and fortunately, H&P somehow manage to do a new edition about every 5 years, often enough to stay current. When the First Edition appeared, it quickly became *the* standard textbook on the topic, to be replaced in that role by each successive edition.

Computer architectures complexify over time, and so do books. The Third Edition was about 1100 pages long. The Fourth has been shrunk back to something more manageable by moving subsidiary details to a CD included with the book.

This edition is well worth having, even if one already has the earlier ones. In particular, the additional material on multiprocessors is especially welcome, given that it has become much more difficult to speed up uniprocessors.

Most people who work in or near computer systems architecture know these books, but I have often recommended them to others, such as technology journalists, venture capitalists, and financial analysts, i.e., people who are rarely computer archtiects, but need to understand computer technology and its trends. Many such have been surprised to find the book was useful to them.

H&P write very clearly, and each chapter outlines its key concepts for a topic, then works down to detailed analyses, and then comes back up to summarize. hence, I've often recommended to people:

1) Read the first few sections of each chapter.

2) In each remaining section per chapter, read until the going gets heavy, then skip to the next section. In some cases, this will happen after reading the first paragraph, but don't worry, the writing will return to a higher level.

3) Read "Concluding Remarks" and any "Fallacies and Pitfalls" or Historical Perspectives" sections at the end of a chapter.

Anyway, I expect this Edition will be just as indispensable as the earlier ones.

I had this book for a graduate course in computer science, and after having been through it, I can say several parts of it are in major need of a rewrite. To be fair, a lot of the material was engaging, and even fun to read; but the layout was so awful that much of what they were saying couldn't be understood by someone not already familiar with the subject without wading through the appendices. In their defense, it is the fourth edition of the book, and the technology has changed much since its inception, so the format is going to be a little messed up. This is also likewise a major strike against them in the area of the exercises: Many of them are very badly written, have little to no relation to what is actually taught in the text, and are fraught with obvious errors that should have been caught two editions ago. On a few of them, even our instructor (who is very knowledgeable) had to throw up his hands and declare that it was anyone's guess.

All in all, this text has a lot good information and potential, but badly needs revision and restructuring if it's to be a truly great learning tool.

I read the 3rd edition cover to cover a few years ago and was amazed by its comprehensive coverage for computing related topics from high level to details. Now, the 4th edition is no worse and worth the money. With significant shift from single processor to multiprocessor, it matches the technology trend and stays on top of current industry challenges. This book is a definitely must have for both learning and referencing.

Regretfully, the publisher decided to save cost by moving many useful appendix chapters to CD and use paperback. The stretch to other topics like virtual memory, networking seems unnecessary. But that doesn't affect the overall book quality.

Combining this architecture level view with in-depth coverage of other important computer areas like "Computer organization and design", "operating system concepts" and "computer network", you will have a clear view of the whole infrastructure of the high tech industry.

For those who have never read this book, I'd emphasize that this is not a computer architecture book that explains some basic stuffs. Instead, this book focuses more on the performance aspect of the computers and I'd say this is the ONLY worthwhile book that discusses from that viewpoint. So, if you are to study the computer performance, your choice is either reading this book or reading tons of papers published by researchers or both. It means you must have a good understanding on the computer architecture before reading this book.

My only complaint is that though the book is available only with paperback now, they didn't reduce the price. It's still worth the price, though.

This book isn't for the timid. It goes deep into several recent CPU designs and explains why the architectures turned out the way they did. There is decent coverage of RISC versus CISC ideas, and why CISC now dominates (hint: it is a combination of luck, marketing, and massive amounts of available transistors, plus new ways of instruction-level parallelism).

It does not cover the absolute latest processors. But it doesn't have to. It will give you the background needed that when you go to the website that have technical details of a new architecture (e.g. Ars Technica), chances are good you will know the concepts they reference.

Who shouldn't buy this: Programmer's in high level languages expecting to learn some black magic way to speed up your code. Even assembly language programmers have been mostly sidelined by the power of a modern CPU to optimize high-level languages.

I've used this book for an Advanced Computer Architecture book in my graduate studies (it seems so has every other graduate course, which has an added bonus).

The book falls in line with most educational books - mediocre at best. The samples are weak and seldom useful. This book is best coupled with classes and online material found via search engine of your choice.

Many universities take the slides and adapt them to their curriculum and post materials accessible online which may be useful.

There are some errors in the book and I have yet to track down the errata (even though they provide a link to a place to find it, seems to be only for professors? Not quite sure what that is about...)

I'm a bit biased, I thought learning about RAID systems, cache optimization, memory optimization, etc - would have been a bit more interesting and challenging. I've mostly found it dull, ancient, and redundant (Zing!).

To read this book, it's best to have covered Computer Organization and Design by HP (the same authors).

Appendices A, B, C, and Chapters 1, 2, and 3 target a 4th year undergraduate/1st year graduate audience. Chapters 4-6 target a continuation course. The beginning material (especially the A/B/C Appendices) forms a very good summary, and sometimes, a more extensive explanation, of some of the fundamental concepts covered in the strictly undergraduate Hennessey-Paterson book 'Computer Organization and Design 3ed' (COD3ed). (COD3ed has now been succeeded by a newer 4ed book). However, even if you are familiar with COD3ed, it can nevertheless prove to be daunting when reading CA4ed. Hence, I recommend the following personally tried reading format for those who are familiar with the undergraduate book (COD3ed) but still want a smooth transition to CA4ed. If you have both books (CA4ed) and (COD3ed)*, try the following sequential reading order:
*please note there is a newer 4th edition of COD on amazon which you might wish to consult or purchase instead of COD3ed

-refresh your memory on computer performance preliminaries in COD3ed
->Chpt4-Performance
->CD sections: "In More Depth-Amdahl's Law" and "In More Depth-MIPS,MOPS,FLOPS"
-read CA4ed's "Chpt1-Fundamentals of Computer Design"
-read CA4ed's Appendices A and B
-read COD3ed's "Chpt7-Large and Fast: Exploiting Memory Hierarchy" and CA4ed's "AppendixC-Review of Memory Hierarchy"
-read Chpts 2 and 3 of CA4ed for dynamic-scheduling/branch prediction/speculation/limitations on ILP
-Chapters 4-6 constitute advanced Memory Hierarchy concepts so essential in multiprocessors. I would recommend rereading some of the basic cache concepts in Appendix C as a refresher, followed by reading Chpts 4 and 5 of CA4ed. Finally reading COD3ed's "Chpt 8-Storage, Networks, and Other Peripherals" followed by CA4's "Chpt 6 Storage Systems".
-The CD includes Appendices on other computer architecture related material: D-Embedded Systems, E-Interconnection Networks, F-Vector Processors, G-Hardware and Software for VLIW and EPIC, H-Large Scale Multiprocessors for Scientific Applications, I-Computer Arithmetic, J-Survey of Instruction Set Architectures, K-Historical Perspectives & References.

Overall, I have to agree with many positive statements that this book sets a new standard for presenting informative material supported by charts/graphs and commentary from years of designer experience. However, one needs to read the material considerably before concluding that some definitions and concepts seem prohibitively unclear (i.e. the branch delay slot, is it an architectural residual from a one-cycle fetch before decoding?). Also, many charts seem redundant and I've been told Hennessey has been known to fudge data (I can't substantiate this). At any rate, sometimes, the author makes the mistake of discussing pitfalls and fallacies at too abstract a level and at great lengths rather than making short, precise, and blunt statements. It seems he is being careful about the academic criticism he may receive and thus makes many ambiguous statements in the process. Towards the end of every chapter, you start feeling there is no structure to the material as all types of loosely defined terms come crashing and conflicting with earlier ones. Many computer science and compiler technology-related issues overlap with computer architecture ones. However, I would hope that an author would consider that it is better to present structured and contained concepts (that capture the spirit of RISC design) and to educate in a piecemeal fashion rather than overwhelm and strike bewilderment. HP slightly evade this approach in their delivery.

What has become clear to me is that microprocessor system-design is a heuristic process and the book sways back and forth between the art and science of design. Ultimately, it is the science and structure we are interested in, as students, but it is the art which dominates the reality, making this a very elusive subject. A more rigorous revision of this book's chart-filled redundancy, lack of structure, poor sentence wording, lack of conciseness and clarity in definitions, may make it a master book. Four stars.

I am pleased with the book I received which is very nice. However, the content is a little too difficult for me. It's very professional.

Information given in chapters is plain text while exercises are purely numeric. Just going through book you cant get down to solving exercises, equations and how to solve questions are not considered important by authors. Even solution manual does not covers any equations and simply starts putting numeric values. If goal of book is to make readers learn concepts this book is good but when it comes to making them solve computer architecture problems, it fares poorly.

The book is a great extension to the computer organization book. Came super quick in the mail and was in perfect condition.