A Primer on Memory Consistency and Cache Coherence, Second Edition
暫譯: 記憶體一致性與快取一致性入門（第二版）

許多現代計算機系統，包括同質和異質架構，都在硬體上支援共享記憶體。在共享記憶體系統中，每個處理器核心都可以讀取和寫入單一的共享地址空間。對於共享記憶體機器，記憶體一致性模型定義了其記憶體系統在架構上可見的行為。一致性定義提供了有關載入和存儲（或記憶體讀取和寫入）及其如何作用於記憶體的規則。作為支援記憶體一致性模型的一部分，許多機器還提供快取一致性協議，以確保多個快取的數據副本保持最新。這本入門書的目標是讓讀者對一致性和一致性有基本的理解。這種理解包括必須解決的問題以及各種解決方案。我們將介紹高層次的概念以及來自真實系統的具體實例。

這第二版反映了自第一版以來十年的進展，並包括除了其他較小的變更外，兩個新章節：一個是關於非 CPU 加速器（重點在 GPU）的一致性和一致性，另一個是指向一致性和一致性的正式工作和工具。

這是一本開放存取的書籍。

Vijay Nagarajan is a Reader at the School of Informatics at the University of Edinburgh. He received a Ph.D. in Computer Science from University of California, Riverside. His research interests span computer architecture, compilers, and computer systems with a focus on memory consistency models and cache coherence protocols. He is a recipient of the Intel early career faculty honour award, a PACT best paper award, and an IEEE Top Picks honorable mention. He has served (or is currently serving) on multiple program committees including ISCA, MICRO, and HPCA. He was General Chair of LCTES 2017 and is currently serving as an Associate Editor of IEEE Computer Architecture Letters (IEEE CAL).Daniel J. Sorin is Professor of Electrical and Computer Engineering and of Computer Science at Duke University. His research interests are in computer architecture, including dependable architectures, verification-aware processor design, and memory system design. He received a Ph.D. and M.S. in electrical and computer engineering from the University of Wisconsin, and he received a BSE in electrical engineering from Duke University. He is the recipient of an NSF Career Award, and he was a Distinguished Visiting Fellow of the Royal Academy of Engineering (UK). He is the Editor-in-Chief of IEEE Computer Architecture Letters, and he is a Founder and Chief Architect of Realtime Robotics, Inc. He is the author of a previous Synthesis Lecture, Fault Tolerant Computer Architecture (2009).Mark D. Hill is John P. Morgridge Professor and Gene M. Amdahl Professor of Computer Sciences at the University of Wisconsin-Madison, where he also has a courtesy appointment in Electrical and Computer Engineering. His research interests and accomplishments are in parallel-computer system design (e.g., data-race-free memory consistency), memory system design (3C model: compulsory, capacity, and conflict misses), and computer simulation (GEMS and gem5). Hill's work is highly collaborative with over 160 co-authors and especially his long-time colleague David A. Wood. He received the 2019 Eckert-Mauchly Award and 2009 ACM SIGARCH Alan Berenbaum Distinguished Service Award. Hill is a fellow of IEEE and the ACM. He served as Chair of the Computer Community Consortium from 2018-2020 and as Wisconsin Computer Sciences Department Chair from 2014-2017. Hill has a Ph.D. in Computer Science from the University of California, Berkeley.David A. Wood is Professor Emeritus of Computer Sciences at the University of Wisconsin, Madison, where he was the Sheldon B. Lubar Chair in Computer Sciences, the Amarand Balinder Sohi Professor in Computer Science, and held a courtesy appointment in Electrical and Computer Engineering. Dr. Wood has a Ph.D. in Computer Science (1990) from UC Berkeley. Dr. Wood is an ACM Fellow (2006), IEEE Fellow (2004), UW Vilas Associate (2011), UW Romnes Fellow (1999), and NSF PYI (1991). Dr. Wood served as Chair of ACM SIGARCH, Area Editor (Computer Systems) of ACM TOMACS, Associate Editor of ACM TACO, Program Committee Chairman of ASPLOS-X (2002), and served on numerous program committees. Dr. Wood has published over 100 technical papers and is an inventor on 19 U.S. patents. Dr. Wood co-led the Wisconsin Wind Tunnel and Wisconsin Multifacet projects with his long-time collaborator Mark D. Hill.

Vijay Nagarajan 是愛丁堡大學資訊學院的讀者。他在加州大學河濱分校獲得計算機科學博士學位。他的研究興趣涵蓋計算機架構、編譯器和計算機系統，專注於記憶體一致性模型和快取一致性協議。他是英特爾早期職業教師榮譽獎的獲得者，還獲得了 PACT 最佳論文獎和 IEEE Top Picks 榮譽提名。他曾在多個程序委員會中任職（或目前仍在任職），包括 ISCA、MICRO 和 HPCA。他是 LCTES 2017 的總主席，並且目前擔任 IEEE 計算機架構快報（IEEE CAL）的副編輯。

Daniel J. Sorin 是杜克大學電機與計算機工程及計算機科學的教授。他的研究興趣在於計算機架構，包括可靠架構、驗證感知的處理器設計和記憶體系統設計。他在威斯康辛大學獲得電機與計算機工程的博士和碩士學位，並在杜克大學獲得電機工程的學士學位。他是 NSF 職業獎的獲得者，並曾擔任英國皇家工程院的傑出訪問學者。他是 IEEE 計算機架構快報的主編，也是 Realtime Robotics, Inc. 的創始人和首席架構師。他是之前 Synthesis Lecture 的作者，題為《容錯計算機架構》（2009）。

Mark D. Hill 是威斯康辛大學麥迪遜分校的 John P. Morgridge 教授和 Gene M. Amdahl 計算機科學教授，他在電機與計算機工程方面也有兼任職位。他的研究興趣和成就包括平行計算機系統設計（例如，無資料競爭的記憶體一致性）、記憶體系統設計（3C 模型：強制性、容量和衝突未命中）以及計算機模擬（GEMS 和 gem5）。Hill 的工作高度合作，擁有超過 160 位共同作者，特別是他的長期同事 David A. Wood。他獲得了 2019 年的 Eckert-Mauchly 獎和 2009 年的 ACM SIGARCH Alan Berenbaum 傑出服務獎。Hill 是 IEEE 和 ACM 的會士。他曾擔任計算機社群聯盟的主席（2018-2020）和威斯康辛計算機科學系主任（2014-2017）。Hill 在加州大學伯克利分校獲得計算機科學博士學位。

David A. Wood 是威斯康辛大學麥迪遜分校的計算機科學榮譽教授，他曾擔任 Sheldon B. Lubar 計算機科學講座教授、Amarand Balinder Sohi 計算機科學教授，並在電機與計算機工程方面擁有兼任職位。Wood 博士在加州大學伯克利分校獲得計算機科學博士學位（1990）。Wood 博士是 ACM 會士（2006）、IEEE 會士（2004）、威斯康辛大學 Vilas 研究員（2011）、威斯康辛大學 Romnes 研究員（1999）和 NSF PYI 獲得者（1991）。Wood 博士曾擔任 ACM SIGARCH 主席、ACM TOMACS 的區域編輯（計算機系統）、ACM TACO 的副編輯、ASPLOS-X（2002）的程序委員會主席，並參與多個程序委員會。Wood 博士已發表超過 100 篇技術論文，並在 19 項美國專利中擔任發明人。Wood 博士與他的長期合作者 Mark D. Hill 共同領導了威斯康辛風洞和威斯康辛多面體項目。