計(jì)算機(jī)系統(tǒng)設(shè)計(jì)原理

內(nèi)容概要

本書由計(jì)算機(jī)系統(tǒng)設(shè)計(jì)的權(quán)威專家、美國麻省理工學(xué)院Saltzer和Kaashoek教授編寫。本書是第一本闡述計(jì)算機(jī)系統(tǒng)設(shè)計(jì)中的基本原理和抽象的教材，是麻省理工開放式課程計(jì)劃(MIT Open Courseware)中“計(jì)算機(jī)系統(tǒng)工程”課程的主教材。計(jì)算機(jī)系統(tǒng)的基本原理橫跨于操作系統(tǒng)、網(wǎng)絡(luò)、數(shù)據(jù)庫、分布式系統(tǒng)、程序設(shè)計(jì)語言、軟件工程以及計(jì)算機(jī)體系結(jié)構(gòu)等方面。通過詳細(xì)分析每個(gè)基本原理的案例，本書演示了如何應(yīng)用這些原理和抽象來解決實(shí)際的計(jì)算機(jī)系統(tǒng)設(shè)計(jì)問題。本書的重點(diǎn)是計(jì)算機(jī)系統(tǒng)的設(shè)計(jì)，因此，本書闡述了那些在實(shí)踐中已證明是成功的抽象概念，如命令、遠(yuǎn)程過程調(diào)用、客戶/服務(wù)器組織結(jié)構(gòu)、文件系統(tǒng)、事務(wù)處理、修復(fù)復(fù)制、讀/寫一致性，以及認(rèn)證與保密消息等概念。這些抽象概念使得計(jì)算機(jī)系統(tǒng)設(shè)計(jì)者可以用功能日益強(qiáng)大的模塊來構(gòu)建計(jì)算機(jī)系統(tǒng)，保護(hù)計(jì)算機(jī)系統(tǒng)避免從意外的編程錯(cuò)誤到惡意入侵在內(nèi)的各種故障。本書描述了如何實(shí)現(xiàn)這些抽象概念，演示了如何把它們應(yīng)用于不同的系統(tǒng)，為讀者日后的設(shè)計(jì)工作打下基礎(chǔ)。

作者簡介

作者：(美國)薩特澤(Jerome H.Saltzer) (美國)卡肖克(M.Frans Kaashoek)Jerome H.Saltzer received the degree of Sc.D.in the field of ElectricaI Engineering in 1 966  from the Massachusetts Institute of Technology.Since 1966 he has been a faculty member in M.I.I.'s Department of ElectricaI Engineering and Computer Science，where he helped to formulate the original undergraduate curriculum in Computer Science.At the M.I.T Computer Science and ArtificiaI Intelligence Laboratory he designed one of the earliest widely-used wo rd..processing systems；he participated in the development of the Multics system，for which he designed the kerneI thread package and with students and colleagues developed the secu rity mechanisms and what would today be known as a microkernel；together with David Clark and David Reed.he articulated the end-to-end argument.a key organizing principle of the Internet；and he was also involved in the design of a token.passing ring IocaI area network.the networking of personal computers，the Kerberos single login authentication system.and digital library systems.Professo r Saltzer was TechnicaI Director of M.I.T Project Athena.a system for undergraduate education and an ea rly example of a system of organization now called"cloud computing".Th roughout his work.he has had a particular interest in the impact of computer systems on privacy and the risks of depending on fragile technology.Professor Saltzer is a member of the NationaI Academy of Engineering.a Fellow of the IEEE and the AAAS，a member of the Association for Computing Machinery.the ACM Committee on Computers and Public Policy.the Catalog Raisonne Schola rs Association，a former member of the Computer Science and TelecommunicatiOns Board of the National Resea rch Council.and a former member of the Mayor's Telecommunications Advisory Board for the City of Newton.Massachusetts.M.Frans Kaashoek received his Ph.D.in 1992 from the Vriie Universiteit in Amsterdam.The Netherlands.fOr his work on g roup communication in the Amoeba distributed operating system，under the supervision of A.S.Tanenbaum.Since 1 993 he has been a facuity member in M.I.T's Department of ElectricaI Enqineering and Compute r Science.where he and Professor Saltzer co-developed the core undergraduate subject on fhe design of computer systems.Professor Kaashoek is a membe r of the Computer Science and Artificial Intelligence Laboratory.whe re his principal field of inte rest is designing and building compute r systems.He co.1eads the paralleI and distributed operating systems group.His past research in collaboration with colleaques and students includes the exokernel operating system，the Click modular router,the RON overlay.the self.certifying file system，the Chord distributed hash table，and the Asbestos secure operating system.He also participated in starting two successfulIT companies.Professor Kaashoek is a member of the NationaI Academy of Engineering and the recipient of several awards.including the inaugural ACM SIGOPS Mark Weiser award for demonstrating creativity and innovation in operating systems research.

書籍目錄

PART IPART II

章節(jié)摘錄

插圖：5.2.2 Sequence Coordination with a Bounded BufferThe implementation with bounded buffers requires coordination between sending and receiving threads because a thread may have to wait until buffer space is available or until a message arrives. Two quite different approaches to thread coordination have developed over the years by researchers in different fields. One approach, usually taken by operating system designers, assumes that the programmer is an all-knowing genius who makes no mistakes. The other approach, usually taken by database designers, assumes that the programmer is a mere mortal, so it provides strong automatic support for coordination correctness, but at some cost in flexibility.The next couple of subsections exhibit the genius approach to coordination, not because it is the best way to tackle coordination problems, but rather to give some intuition about why it requires a coordination genius, and thus should be subcontracted to such a specialist whenever possible. In addition, to implement the database approach the designer of the automatic coordination support approach must use the genius approach. Chapter 9 [on-line] uses the concepts introduced in this chapter to implement the database approach for mere mortals.

編輯推薦

《計(jì)算機(jī)系統(tǒng)設(shè)計(jì)原理(影印版)》：大學(xué)計(jì)算機(jī)教育國外著名教材系列。

圖書封面

圖書標(biāo)簽Tags

無

評(píng)論、評(píng)分、閱讀與下載

---

    計(jì)算機(jī)系統(tǒng)設(shè)計(jì)原理 PDF格式下載

---