The first three editions ofStructured Computer Orgunization were based on the idea that a computer can be regarded as a hierarchy of levels, each one performing some well-defined function. This fundamental concept is as valid today as it was when the first edition came out, so it has been retained as the basis for the foudh edition. As in the first three editions, the digital logic level, the micro-architecture level, the instruction set architecture level, the operating system machine level, and the assembly language level are all discussed in detail (although we have changed some of the names to refiect modem practice).
The fourth edition of this best-selling introduction to computer hardware and architecture has been heavily revised to reflect the latest changes in the rapidly changing computer industry. Professor Tanenbaum has maintained his popular method ofpresenting the computer as a hierarchy of levels, each with a well-defined filnction. The book is written in a style and level ofdetail that covers all the major areas, but is still accessible to a broad range of readers.
The first three editions of this book were based on the idea that a computer can be regarded as a bierarchy of levels, each one performing some well-defined function. This fundamental concept is as valid today as it was when the first edition came out, so it has been retained as the basis for the fourth edition. As in the first three editions, the digital logic level, the microarchitecture level, the insvuction set architecture level, the operating system machine level, and the assembly language level are all discussed in detail (although we have changed some of the names to reaect modem practice).
Although the basic structure has been maintained, this fourth edition contains many changes, both small and large, that bring it up to date in the rapidly changing computer industry. For example, all the code examples, which were in Pascal,have been rewritten in Java. renecting the popularity of Java in the computer world. Also, the example machines used have been broaght up to date. The current examples are the Intel Pentium II. the Sun UltraSPARC II, and the Sun picoJava II, an embedded low-cost hardware Java chip.
Multiprocessors and parallel computers have also come in widespread use since the third edition, so the material on parallel architectures has been completely redone and greatly expanded, now covering a wide range of topics, from multiprocessors to COWs.
The book has become longer over the years (although still not as long as some other popular books on the subject). Such an expansion is inevitable as a subject develops and there is more known about it. As a result. when the book is used for a course, it may not always be possible to finish the book in a single course (e.g..in a trimester system). A possible approach would be to do all of Chaps. l, 2, and 3, the first part of Chap. 4 (up through and including Sec. 4.4), and Chap. 5 as a bare minimum. The remaining time could be filled with the rest of Chap. 4, and parts of Chaps. 6, 7, and 8, depending on the interest of the instructor.
A chapter-by-chapter rundown of the major changes since the third edition follows. Chapter I still contains an historical overview of computer architecture,pointing out how we got where we are now and what the milestones were along the way. The enlarged spectrum of computers that exist is now discussed, and ourthree major examples (PenLium II, UltraSPARC II, and picoJava II) are introduced.
In Chapter 2, the material on input/output devices has been updated,emphasizing the technology of modem devices, including RAID dislcs, CDRecordables, DVD, and color printers, among many others.
Chapter 3 (digital logic level) has undergone some revision and now treats computer buses and modem I/O chips. The major change here is additional material on buses, especially the PCI bus and the USB bus. The three new examples are described here at the chip level.
Chapter 4 (now called the microarchitecture level) has been completely rewritten. The idea of using a detailed example of a microprogrammed machine to illustrate the ideas of data path control has been retained, bat the example machine is now a subset of the Java Vinual Machine. The underlying microarchitecture has been comespondingly changed. Several iterations of the design are given, sbowing what trade-offs are possible in terms of cost and performance.
The last example, the Mic-4, uses a seven-stage pipeline and provides an easy inuoduction to how imponant modem computers, such as the Pentium II, work.A new section on improving performance has been added, focusing on the most recent techniques such as caching, branch prediction, (superscalar) out-of-order execution, specolative execution, and predication. The new example machines are discussed at the microarchitecture level.
Chapter 5 (now called the instruction set architecture level) deals with what many people refer to as “machine language.” The Pentium II, UltraSPARC II and Java Vinual Machine are used as the primary examples here.
Chapter 6 (operating system machine level) has examples for the Pentium II (Windows NT) and UltraSPARC II (UNIX). The former is new and has many features that are wonh looking at, but UNIX is still a reliable workhorse at many universities and companies and is well worth examining in detail as well due to its simple and elegant design.
Chapter 7 (assembly language level) has been brought up to date by using examples from the machines we have been studying. New material on dynamic linking has been added as well.
Chapter 8 (parallel computer architectures) has been completely rewritten from the third edition. It now covers both multiprocessors (UMA, NUMA, and COMA) in detail, as well as multicomputers (MPP and COW).
The bibliography has been extensively revised and brought up to date.Well over two-thirds the references refer to works published after the third edition was published. Binary numbers and noating-point numbers have not undergone much change recently, so the appendices are largely the same as in the
Finally, some problems have been revised and many new problems have been added since the third edition. Accordingly, a new problem solutions manual is available from Prentice Hall. It is available only to faculty members. who can request a free copy from their Prentice Hall representative.
A Web site for this book is available. PostScript ales ror all the illusuations used in the book are available electronically. They can be fetched and printed. for example. for making overhead sbeets. In addition, a simulator and other and software tools are there too. The URL for this site is http://www. cs. vu.nl/～ast/sco4/
The simulator and software tools were produced by Ray Ontko. The author wishes to express his gratitude to Ray for producing these extremely useful programs.
A number of people kave read (parts Of) the manuscript and provided useful suggestions or have been helpful in other ways. In particular, I would like to alank Henri Bal, Alan Charlesworth, Kourosh Gharachorloo, Marcus Goncalves,Karen Panetta Lentz, Timothy Mattson, Harlan McOhan, Miles Murdocca, Kevin Normoyle, Mike O'Connor, Mitsunori Ogihara, Ray Ontko, Aske Plaat, William Potvin II, Nagarajan Prabhakaran. James H. Pugsley, Ronald N. Schroeder, Ryan Shoemaker, Charles Silio, Jr., and Dale Skrien for their help, for which I am most grateful. My students, especially Adriaan Bon, Laura de Vries, Dolf Loth, Guido van 't Noordende, have also helped debug the text. Thank you.
I would especially like to thank Jim Goodman for his many contributions to this book, especially to Chaps. 4 and 5. The idea of using the Java Vinual Machine was his, as were the microarchitectures for implementing it. Many of the advanced ideas were due to him. The book is far better for his having put in so much effort.
Finally, I would like to thank Suzanne for her patience for my long hours in .front of my Pentium. From my point of view the Pentium is a big improvenlent over my older 386 but from hers, it does not make much difference. I also want to thank Barbara and Marvin for being great kids and Bram for always being quiet when I was Uying to write.
Andrew S. Tanenbaum
Andrew S.Tanenbaum：Andrew S．Tanenbaum获得过美国麻省理工学院的理学学士学位和加利福尼亚大学伯克利分校的哲学博士学位，目前是荷兰阿姆斯特丹Vrije大学的计算机科学系的教授，并领导着一个计算机系统的研究小组。同时，他还是一家计算与图像处理学院的院长，这是由几家大学合作成立的研究生院。尽管社会工作很多，但他并没有中断学术研究。多年来，他在编译技术、操作系统、网络及局域分布式系统方面进行了大量的研究工作。目前的主要研究方向是设计规模达数百万用户的广域分布式系统。在进行这些研究项目的基础上，他在各种学术杂志及会议上发表了70多篇论文。他同时还是5本计算机专著的作者。 Tanenbaum教授还开发了大量的软件。他是Amsterdan编译器的原理设计师，这是一个被广泛使用的,用来编写可移植编译器的工具箱。他领导编写的MINIX，是一个用于操作系统教学的类UNIX的小型操作系统。他和他的博士研究生及其他编程人员一道设计的Amoeba分布式操作系统，是一个高性能的微内核分布式操作系统。目前，可在因特网上免费得到MINIX及Amoeba，用于教学和研究。他的一些博士研究生，在获得学位后继续进行研究，并取得了更大的成就，赢得了社会的赞誉，对此他深感自豪。人们称他为桃李满天下的教育家。
1 INTRODUCTION
l.l STRUCTURED COMPUTER ORGANIZATION
l.2 MILESTONES IN COMPUTER ARCHITE
l.3 THE COMPUTER ZOO
l.4 EXAMPLE COMPUTER FAMILIES
l.5 OUTLINE OF THIS BOOK
2 COMPUTER SYSTEMS ORGANIZATION
2.1 PROCESSORS
2.2 PRIMARY MEMORY
2.3 SECONDARY MEMORY
2.4 INPUT/OUTPUT
2.5 SUMMARY
3 THE DIGITAL LOGIC LEVEL
3.I GATES AND BOOLEAN ALGEBRA
3.2 BASIC DIGITAL LOGIC CIRCUITS
3.3 MEMORY
3.4 CPU CHIPS AND BUSES
3.5 EXAMPLE CPU CHIPS
3.6 EXAMPLE BUSES
3.7 INTERFACING
3.8 SUMMARY
4 THE MICROARCHtTECTURE LEVEL
4.l AN EXAMPLE MICROARCHITECTURE
4.2 AN EXAMPLE ISA: IJVM
4.3 AN EXAMPLE IMPLEMENTATION
4.4 DESIGN OF rHE MICROARCHITECTURE LEVEL
4.5 IMPROVING PERFORMANCE
4.6 EXAMPLES OF THE MICROARCHITECTURE LEVEL
4.7 SUMMARY
5 THE INSTRUCTION SET ARCHITECTURE LEVEL
s.l OVERVIEW OF THE ISA LEVEL
5.2 DATA TYPES
5.3 INSTRUCTION FORMATS
5.4 ADDRESSING
5.5 INSTRUCTION TYPES
5.6 FLOW OF CONTROL
5.7 A DETAILED EXAMPLE: THE TOWERS OF HANOI
5.8 THE INTEL IA-64
5.9 SUMMARY
6 THE OPERATING SYSTEM MACHINE LEVEL
6.l VIRTUAL MEMORY
6.2 VIRTUAL UO INSTRUCTIONS
6.3 VIRTUAL INSTRUCTIONS FOR PARALLEL PROCESSING
6.4 EXAMPLE OPERATING SYSTZMS
6.5 SUMMARY
7 THE ASSEMBLY LANGUAGE LEVEL I
7.I INTRODUCTION TO ASSEMBLY LANGUAGE
7.2 MACROS
7.3 THE ASSEMBLY PROCESS
7.4 LINKINO AND LOADING
7.5 SUMMARY
8 PARALLEL COMPUTER ARCHITECTURES
8.l DESIGN ISSUES FOR PARALLEL COMPUTERS
8.2 SIMD COMPUTERS
8.3 SHARED-MEMORY MULTIPROCESSORS
8.4 MESSAGE-PASSING MULTICOMPUTERS
8.5 SUMMARY
9 READING LIST AND BIBLIOeRAPHY
9. I SUGGESTIONS FOR FURTHER READING
9.2 ALPHABETICAL BIBLIOORAPHY
A BINARY NUMBERS
A.I FINITE-PRECISION NUMBERS
A.2 RADIX NUMBER SYSTEMS
A.3 CONVERSION FROM ONE RADIX TO ANOTHER
A.4 NEGATIVE BINARY NUMBERS
A.5 BINARY ARnHMETIC
B FLOATING－POINT NUMBERS
B.l PRINCIPLES OF FLOATING POINT
B.2 IEEE FLOATING-POINT STANDARD
INDEX
