Computer Architecture

John L. Hennessy

出版时间

2011-09-30

ISBN

9780123838728

评分

★★★★★
书籍介绍

The computing world today is in the middle of a revolution: mobile clients and cloud computing have emerged as the dominant paradigms driving programming and hardware innovation today. The Fifth Edition of Computer Architecture focuses on this dramatic shift, exploring the ways in which software and technology in the cloud are accessed by cell phones, tablets, laptops, and other mobile computing devices. Each chapter includes two real-world examples, one mobile and one datacenter, to illustrate this revolutionary change. Updated to cover the mobile computing revolution Emphasizes the two most important topics in architecture today: memory hierarchy and parallelism in all its forms. Develops common themes throughout each chapter: power, performance, cost, dependability, protection, programming models, and emerging trends ("What's Next") Includes three review appendices in the printed text. Additional reference appendices are available online. Includes updated Case Studies and completely new exercises.

John L. Hennessy is the president of Stanford University, where he has been a member of the faculty since 1977 in the departments of electrical engineering and computer science. Hennessy is a fellow of the IEEE and the ACM, a member of the National Academy of Engineering, the National Academy of Science, the American Academy of Arts and Sciences, and the Spanish Royal Academy o...

(展开全部)

AI导读
核心看点
  • 本书第五版全面转向移动计算与云计算范式,深入剖析移动端设备与数据中心在架构设计上的根本差异,强调软件定义硬件及云边协同的技术趋势,反映当前计算领域的革命性变革。
  • 全书贯穿性能、功耗、成本、可靠性、保护及可编程性等核心设计权衡原则,通过定量分析方法,指导读者理解如何在复杂约束下优化系统架构,培养严谨的工程思维与设计伦理。
  • 系统讲解内存层次结构、指令级并行、数据级并行及线程级并行等核心技术,涵盖GPU、多处理器及大规模仓库级计算机设计,提供从底层硬件到上层系统的全景式架构知识体系。
适合谁读
  • 计算机体系结构领域的专业研究人员、高校教师及研究生,旨在深入理解并行计算、内存一致性模型及大规模分布式系统底层原理,进行前沿技术探索与学术研究。
  • 从事高性能计算、云计算基础设施、嵌入式系统或底层驱动开发的资深工程师,需要掌握硬件加速、功耗优化及大规模集群部署的技术细节,以提升系统整体效能与稳定性。
  • 已具备操作系统、编译原理及计算机组成原理基础,希望进阶学习系统级性能分析与优化,弥补硬件知识短板,提升对复杂系统整体架构理解能力的计算机科学与技术专业高年级学生。
读前提醒
  • 严禁直接阅读中文版,翻译质量极差且逻辑混乱,严重误导读者。必须使用英文原版,以确保准确理解专业术语与复杂的技术逻辑,避免被劣质译文浪费宝贵时间与精力。
  • 本书内容极其艰深,不适合初学者。必须先精通《计算机组成与设计》及《深入理解计算机系统》等基础教材,具备扎实的硬件与系统基础,否则将因缺乏前置知识而难以理解定量分析部分。
  • 不要试图通读全书,应根据自身研究方向选择性阅读。附录部分包含大量基础硬件知识,适合基础薄弱者补充;正文聚焦高级架构设计,需结合具体项目需求,重点研读并行计算与数据中心相关章节。
读者共识
  • 本书是计算机体系结构领域的绝对圣经,内容权威且极具深度,但难度极高,被公认为计算机专业学生的终极挑战之一,许多读者表示阅读过程痛苦但收获巨大,对提升系统观有决定性帮助。
  • 读者普遍反映书中关于并行计算、GPU架构及大规模数据中心的内容极具前瞻性,对理解现代云计算与AI硬件基础至关重要,但同时也指出其内容过于理论化,与日常应用开发距离较远。
  • 尽管本书评价极高,但多数读者建议仅作为参考书查阅,而非从头到尾精读。因其内容更新快且难度陡增,更推荐结合具体技术栈需求,利用其附录补充基础知识,避免陷入无谓的学术深坑。

本导读基于书籍简介、目录、原文摘录、短评和书评生成,不等同于全文精读。

精彩摘录
  • "Measuring performance of multiprocessors by linear speedup versus execution time."
  • "A natural question is whether WSCs are similar to modern clusters for high-performance computing. Although some have similar scale and cost..."
  • "The HPC clusters also tend to have long-running jobs that keep the server fully utilized, even for weeks at a time, while the utilization of servers in WSCs ranges between 10% and 50% and varies every day."
  • "How do WSCs compare to conventional datacenters?"
  • "The Google rule of thumb is currently to use the low-end range of server class computers."
  • "在设计上必须有所取舍时,一定要优先考虑较常发生的事件"
  • "1.10 Putting It All Together: Performance, Price, and Power personal mobile devices (PMDs)"
  • "(2) 存储器寻址。80x86(见附录图A-2)不需要对齐,但如果操作数是对齐的,访问速度通常会更快一些。 (See Figure A.5 on page A-8.) The 80x86 does not require alignment, but accesses are generally faster if operands are aligned. (见附录A图A.5)"
作者简介
John L. Hennessy is the president of Stanford University, where he has been a member of the faculty since 1977 in the departments of electrical engineering and computer science. Hennessy is a fellow of the IEEE and the ACM, a member of the National Academy of Engineering, the National Academy of Science, the American Academy of Arts and Sciences, and the Spanish Royal Academy of Engineering. He received the 2001 Eckert-Mauchly Award for his contributions to RISC technology, the 2001 Seymour Cray Computer Engineering Award, and shared the John von Neumann award in 2000 with David Patterson. After completing the project in 1984, he took a one-year leave from the university to co-found MIPS Computer Systems, which developed one of the first commercial RISC microprocessors. After being acquired by Silicon Graphics in 1991, MIPS Technologies became an independent company in 1998, focusing on microprocessors for the embedded marketplace. As of 2004, over 300 million MIPS microprocessors have been shipped in devices ranging from video games and palmtop computers to laser printers and network switches. Hennessy's more recent research at Stanford focuses on the area of designing and exploiting multiprocessors. He helped lead the design of the DASH multiprocessor architecture, the first distributed shared-memory multiprocessors supporting cache coherency, and the basis for several commercial multiprocessor designs, including the Silicon Graphics Origin multiprocessors. Since becoming president of Stanford, revising and updating this text and the more advanced Computer Architecture: A Quantitative Approach has become a primary form of recreation and relaxation. David A. Patterson was the first in his family to graduate from college (1969 A.B UCLA), and he enjoyed it so much that he didn't stop until a PhD, (1976 UCLA). After 4 years developing a wafer-scale computer at Hughes Aircraft, he joined U.C. Berkeley in 1977. He spent 1979 at DEC working on the VAX minicomputer. He and colleagues later developed the Reduced Instruction Set Computer (RISC). By joining forces with IBM's 801 and Stanford's MIPS projects, RISC became widespread. In 1984 Sun Microsystems recruited him to start the SPARC architecture. In 1987, Patterson and colleagues wondered if tried building dependable storage systems from the new PC disks. This led to the popular Redundant Array of Inexpensive Disks (RAID). He spent 1989 working on the CM-5 supercomputer. Patterson and colleagues later tried building a supercomputer using standard desktop computers and switches. The resulting Network of Workstations (NOW) project led to cluster technology used by many startups. He is now working on the Recovery Oriented Computing (ROC) project. In the past, he served as Chair of Berkeley's CS Division, Chair and CRA. He is currently serving on the IT advisory committee to the U.S. President and has just been elected President of the ACM. All this resulted in 150 papers, 5 books, and the following honors, some shared with friends: election to the National Academy of Engineering; from the University of California: Outstanding Alumnus Award (UCLA Computer Science Department), McEntyre Award for Excellence in Teaching (Berkeley Computer Science), Distinguished Teaching Award (Berkeley); from ACM: fellow, SIGMOD Test of Time Award, Karlstrom Outstanding Educator Award; from IEEE: fellow, Johnson Information Storage Award, Undergraduate Teaching Award, Mulligan Education Medal, and von Neumann Medal.
目录
Table of Contents
Printed Text
Chap 1: Fundamentals of Quantitative Design and Analysis
Chap 2: Memory Hierarchy Design
Chap 3: Instruction-Level Parallelism and Its Exploitation

显示全部
用户评论
传说中每翻一遍都能有新收获的书: 从低到高的视角思考并行 1. instruction level parallelism 2. thread level parallelism 3. data level parallelism 4. request level parallelism
体系教材 ------ 结课标注
关于并行运算讲得挺多,我因此也走入了CUDA的世界
果然经典!
血和泪...
Coursera的Computer Architecture参考书目 - 在面试被考了一次Architecture挂了之后痛下决心把它学了。因为缺乏硬件基础读的磕磕绊绊,然而收获颇丰,感觉补上了操作系统+编译原理之后缺少的一块内容体系。
CSAPP进阶,很赞 有深度,有时间把附录材料刷一遍,加油💪
上课的课本,书是好书,可惜我太菜。以后有时间应该仔细重读。
收藏