基本信息
- 原书名：Distributed Operating Systems
- 原出版社： Prentice Hall/Pearson
编辑推荐
本书作者Tanenbaum教授是国际知名的计算机科学专家，是计算机操作系统、分布式操作系统和计算机网络领域的大师。本书是其三部操作系统名著之一，另外两部是《现代操作系统》和《操作系统：设计和实现》。这三本教材在操作系统领域享有极高的声誉，被国内外众多大学广为采用。
内容简介
书籍
计算机书籍
本书作者Tanenbaum教授是国际知名的计算机科学专家，是计算机操作系统、分布式操作系统和计算机网络领域的大师。本书是其三部操作系统名著之一，另外两部是《现代操作系统》和《操作系统：设计和实现》。这三本教材在操作系统领域享有极高的声誉，被国内外众多大学广为采用。.
本书全面而详尽地阐述了分布式操作系统，涵盖分布式操作系统的方方面面，并重点介绍设计和构造分布式操作系统的基本原理、算法和实现技术。书中前半部分较全面地介绍了分布式系统环境中的通信、同步、进程、文件系统和存储器管理等方面的概念与算法；后半部分详细地给出了分布式系统的4个实例：Amoeba、Mach、Chorus和OSF/DCE，介绍其设计思想和实现技术。..
本书的特点是理论与实践紧密结合，既注重基础知识的系统性和先进性，同时又兼顾实践环节。因此，本书的适用对象非常广泛，既可作为本科生、研究生的分布式操作系统课程的教学用书，也可供从事分布式计算研究和工程应用的科研人员与工程技术人员参考。...
计算机书籍
本书作者Tanenbaum教授是国际知名的计算机科学专家，是计算机操作系统、分布式操作系统和计算机网络领域的大师。本书是其三部操作系统名著之一，另外两部是《现代操作系统》和《操作系统：设计和实现》。这三本教材在操作系统领域享有极高的声誉，被国内外众多大学广为采用。.
本书全面而详尽地阐述了分布式操作系统，涵盖分布式操作系统的方方面面，并重点介绍设计和构造分布式操作系统的基本原理、算法和实现技术。书中前半部分较全面地介绍了分布式系统环境中的通信、同步、进程、文件系统和存储器管理等方面的概念与算法；后半部分详细地给出了分布式系统的4个实例：Amoeba、Mach、Chorus和OSF/DCE，介绍其设计思想和实现技术。..
本书的特点是理论与实践紧密结合，既注重基础知识的系统性和先进性，同时又兼顾实践环节。因此，本书的适用对象非常广泛，既可作为本科生、研究生的分布式操作系统课程的教学用书，也可供从事分布式计算研究和工程应用的科研人员与工程技术人员参考。...
作译者
Andrew S. Tanenbaum 拥有美国麻省理工学院的理学学士学位和加州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教授，并领导着一个计算机系统的研究小组。多年来，他在操作系统、编译技术、网络及周域分布式系统方面进行了大量的研究工作，并在各种学术杂志及会议上发表了多篇论文，同时还是5本计算机专著的作者。Tanenbaum是ACM会员、IEEE资深会员、荷兰皇家艺术和科学学院院士，多次获得计算机教育杰出贡献奖。他还入选了《世界名人录》。...
目录
PREFACE .
1 INTRODUCTION TO DISTRIBUTED SYSTEMS
1.1 WHAT IS A DISTRIBUTED SYSTEM?
1.2 GOALS
1.3 HARDWARE CONCEPTS
1.4 SOFTWARE CONCEPTS
1.5 DESIGN ISSUES
1.6 SUMMARY
2 COMMUNICATION IN DISTRIBUTED SYSTEMS
2.1 LAYERED PROTOCOLS
2.2 ASYNCHRONOUS TRANSFER MODE NETWORKS
2.3 THE CLIENT-SERVER MODEL
2.4 REMOTE PROCEDURE CALL
2.5 GROUP COMMUNICATION
2.6 SUMMARY
3 SYNCHRONIZATION IN DISTRIBUTED SYSTEMS
3.1 CLOCK SYNCHRONIZATION
3.2 MUTUAL EXCLUSION
3.3 ELECTION ALGORITHMS
3.4 ATOMIC TRANSACTIONS
1 INTRODUCTION TO DISTRIBUTED SYSTEMS
1.1 WHAT IS A DISTRIBUTED SYSTEM?
1.2 GOALS
1.3 HARDWARE CONCEPTS
1.4 SOFTWARE CONCEPTS
1.5 DESIGN ISSUES
1.6 SUMMARY
2 COMMUNICATION IN DISTRIBUTED SYSTEMS
2.1 LAYERED PROTOCOLS
2.2 ASYNCHRONOUS TRANSFER MODE NETWORKS
2.3 THE CLIENT-SERVER MODEL
2.4 REMOTE PROCEDURE CALL
2.5 GROUP COMMUNICATION
2.6 SUMMARY
3 SYNCHRONIZATION IN DISTRIBUTED SYSTEMS
3.1 CLOCK SYNCHRONIZATION
3.2 MUTUAL EXCLUSION
3.3 ELECTION ALGORITHMS
3.4 ATOMIC TRANSACTIONS
前言
With the publication of Distributed Operating Systems I have now completed my trilogy on operating systems. The three volumes of this trilogy are: .
· Operating Systems: Design and Implementation
· Distributed Operating Systems
· Modern Operating Systems
The three volumes are not completely independent, however. For schools having a two-course sequence in operating systems (or an undergraduate course plus a graduate course), one possible choice is to use Operating Systems: Design and Implementation in the first course and Distributed Operating Systems in the second one.
The former book treats the standard principles of single-processor systems, including processes, synchronization, I/O, deadlocks, memory management, file systems, security, and so on. It also illustrates these principles in great detail through the use of MINIX, a UNIX-clone whose source listing is given in an appendix. MINIX is available on diskette from Prentice Hall for the IBM PC (8088 and up), Atari, Amiga, Macintosh, and SPARC processors.
The latter book (this one), covers distributed operating systems in detail, including communication, synchronization, processes, file systems, and memory management, but this time in the context of distributed systems. Four examples of distributed systems are given in great detail: Amoeba, Mach, Chorus, and DCE. Amoeba is available for free to universities for educational use. It runs on the Intel 386/486, SPARC, and Sun 3 processors. For information on how to obtain Amoeba please FTP the file amoeba/Intro.ps.Z from ftp.cs.vu.nl or contact the author by electronic mail at ast@cs.vu.nl. Potential users should be forewarned that Amoeba is considerably more complex than MINIX: the documentation alone (available by FTP), runs to well over 1000 pages and the system requires at least five large machines and an Ethernet to run well.
By studying these two books in sequence and using both MINIX and Amoeba, students will obtain a thorough knowledge of the principles and practice of both single-processor and distributed operating systems. Now that the trilogy is completed, I plan to revise MINIX and the book describing it. ..
For universities or computer professionals with less time available, Modern Operating Systems can be thought of as a condensed version of the other two books. It provides an introduction to the principles of both single-processor and distributed operating systems, but without the detailed example of MINIX. It also omits many of the advanced topics present in this book, including an introduction to ATM, fault-tolerant distributed systems, real time distributed systems, distributed shared memory, Chorus, DCE, and other topics. In all, about 230 pages of material on distributed systems present in this book have been omitted from Modern Operating Systems.
Many people have helped me with this book. I would especially like to thank the following people for reading portions of the manuscript and giving me many useful suggestions for improvement: Irina Athanasiu, Henri Bal, saniya Ben Hassen, David Black, John Carter, Randall Dean, Wiebren de Jonge, John Dugas, Dick Grune, Anoop Gupta, Frans Kaashoek, Marcus Koebler, Hermann Kopetz, Ed Lazowska, Dan Lenoski, Kai Li, Marc Maathuis, David Mosberger, Douglas Orr, Craig Partridge, Carlton Pu, Marc Rozier, Rich Salz, Mike Schroeder, Karsten Schwan, Greg Sharp, Dennis Shasha, Sol Shatz, Jennifer Steiner, Chuck Thacker, John Turek, Walt Tuvell, Leendert van Doom, Robbert van Renesse, Kees Verstoep, Ellen Zegura, Willy Zwaenpoel, and the anonymous reviewers. My editor, Bill Zobrist, put up with my attempts to get everything perfect with nary a whimper.
Despite all this help, no doubt some errors remain. That seems to be inevitable, no matter how many people read the manuscript. People who wish to report errors should contact me by electronic mail.
Finally, I would like to thank Suzanne again. After eight books, she knows the implications of another one, but her patience and love are boundless. I also want to thank Barbara and Marvin for using their computers and leaving mine alone (except for the printer). Teaching them how to use PC word processing programs has made me appreciate troff more than ever. Finally, I would like to thank Little Bram for being quiet while I was writing. ...
Andrew S. Tanenbaum
· Operating Systems: Design and Implementation
· Distributed Operating Systems
· Modern Operating Systems
The three volumes are not completely independent, however. For schools having a two-course sequence in operating systems (or an undergraduate course plus a graduate course), one possible choice is to use Operating Systems: Design and Implementation in the first course and Distributed Operating Systems in the second one.
The former book treats the standard principles of single-processor systems, including processes, synchronization, I/O, deadlocks, memory management, file systems, security, and so on. It also illustrates these principles in great detail through the use of MINIX, a UNIX-clone whose source listing is given in an appendix. MINIX is available on diskette from Prentice Hall for the IBM PC (8088 and up), Atari, Amiga, Macintosh, and SPARC processors.
The latter book (this one), covers distributed operating systems in detail, including communication, synchronization, processes, file systems, and memory management, but this time in the context of distributed systems. Four examples of distributed systems are given in great detail: Amoeba, Mach, Chorus, and DCE. Amoeba is available for free to universities for educational use. It runs on the Intel 386/486, SPARC, and Sun 3 processors. For information on how to obtain Amoeba please FTP the file amoeba/Intro.ps.Z from ftp.cs.vu.nl or contact the author by electronic mail at ast@cs.vu.nl. Potential users should be forewarned that Amoeba is considerably more complex than MINIX: the documentation alone (available by FTP), runs to well over 1000 pages and the system requires at least five large machines and an Ethernet to run well.
By studying these two books in sequence and using both MINIX and Amoeba, students will obtain a thorough knowledge of the principles and practice of both single-processor and distributed operating systems. Now that the trilogy is completed, I plan to revise MINIX and the book describing it. ..
For universities or computer professionals with less time available, Modern Operating Systems can be thought of as a condensed version of the other two books. It provides an introduction to the principles of both single-processor and distributed operating systems, but without the detailed example of MINIX. It also omits many of the advanced topics present in this book, including an introduction to ATM, fault-tolerant distributed systems, real time distributed systems, distributed shared memory, Chorus, DCE, and other topics. In all, about 230 pages of material on distributed systems present in this book have been omitted from Modern Operating Systems.
Many people have helped me with this book. I would especially like to thank the following people for reading portions of the manuscript and giving me many useful suggestions for improvement: Irina Athanasiu, Henri Bal, saniya Ben Hassen, David Black, John Carter, Randall Dean, Wiebren de Jonge, John Dugas, Dick Grune, Anoop Gupta, Frans Kaashoek, Marcus Koebler, Hermann Kopetz, Ed Lazowska, Dan Lenoski, Kai Li, Marc Maathuis, David Mosberger, Douglas Orr, Craig Partridge, Carlton Pu, Marc Rozier, Rich Salz, Mike Schroeder, Karsten Schwan, Greg Sharp, Dennis Shasha, Sol Shatz, Jennifer Steiner, Chuck Thacker, John Turek, Walt Tuvell, Leendert van Doom, Robbert van Renesse, Kees Verstoep, Ellen Zegura, Willy Zwaenpoel, and the anonymous reviewers. My editor, Bill Zobrist, put up with my attempts to get everything perfect with nary a whimper.
Despite all this help, no doubt some errors remain. That seems to be inevitable, no matter how many people read the manuscript. People who wish to report errors should contact me by electronic mail.
Finally, I would like to thank Suzanne again. After eight books, she knows the implications of another one, but her patience and love are boundless. I also want to thank Barbara and Marvin for using their computers and leaving mine alone (except for the printer). Teaching them how to use PC word processing programs has made me appreciate troff more than ever. Finally, I would like to thank Little Bram for being quiet while I was writing. ...
Andrew S. Tanenbaum
