数据库系统实现(英文影印版·第2版)

书籍
计算机书籍
　　本书是关于数据库系统实现方面内容最为全面的著作之一，是美国斯坦福大学计算机科学专业数据库系列课程第二门课程的指定教材。书中从数据库实现者的角度对数据库系统实现原理进行了深入阐述，并具体讨论了数据库管理系统的三个主要成分——存储管理器、查询处理器和事务管理器的实现技术。斯坦福大学计算机科学专业数据库系列课程第一门课程的内容包括数据库设计和数据库编程，本书的后两位作者jeffrey d. ullman和jennifer widom为该课程编写的教材《数据库系统基础教程》（a first course in database systems）第3版的中文翻译版和英文影印版已由机械工业出版社出版。.
　　 本书内容深入且全面，技术实用且先进，叙述深入浅出，是一本难得的高层次的教材，适合作为高等院校计算机专业研究生的教材或本科生的教学参考书，也适合作为从事相关研究或开发工作的专业技术人员的高级参考资料。...

1 the worlds of database systems.
i relational database modeling
2 the relational model of data
3 design theory for relational databases
4 high-level database models
ii relational database programming
5 algebraic and logical query languages
6 the database language sql
7 constraints and triggers
8 views and indexes
9 sql in a server environment
10 advanced topics in relational database
iii modeling and programming for semistructured data
11 the semistructured-data model
12 programming languages for xml ..
iv database system implementation
13 secondary storage management
14 index structures
15 query execution
16 the query compiler

　　This book covers the core of the material taught in the database seauence at Stanford. The introductory course, CS145, uses the first twelve chapters, and is designed for all students those who want to use database systems as well as those who want to get involved in database implementation. The second course, CS245 on database implementation, covers most of the rest of the book. However, some material is covered in more detail in special topics courses. These include CS346 (implementation project), which concentrates on query optimization as in Chapters 15 and 16. Also, CS345A, on data mining and Web mining, covers the material in the last two chapters. .
　　What's New in the Second Edition
　　After a brief introduction in Chapter 1, we cover relational modeling in Chapters 2-4. Chapter 4 is devoted to high-level modeling. There, in addition to the E/R model, we now cover UML (Unified Modeling Language). We also have moved to Chapter 4 a shorter version of the material on ODL, treating it as a design language for relational database schemas.
　　The material on functional and multivalued dependencies has been modified and remains in Chapter 3. We have changed our viewpoint, so that a functional dependency is assumed to have a set of attributes on the right. We have also given explicitly certain algorithms, including the "chase," that allow us to manipulate dependencies. We have augmented our discussion of third normal form to include the 3NF synthesis algorithm and to make clear what the tradeoff between 3NF and BCNF is.
　　Chapter 5 contains the coverage of relational algebra from the previous edition, and is joined by (part of) the treatment of Datalog from the old Chapter 10. The discussion of recursion in Datalog is either moved to the book's Web site or combined with the treatment of recursive SQL in Chapter 10 of this edition.
　　Chapters 6-10 are devoted to aspects of SQL programming, and they represent a reorganization and augmentation of the earlier book's Chapters 6, 7, 8, and parts of 10. The material on views and indexes has been moved to its own chapter, number 8, and this material has been augmented with a discussion of important new topics, including materialized views, and automatic selection of indexes.
　　The new Chapter 9 is based on the old Chapter 8 (embedded SQL). It is introduced by a new section on 3-tier architecture. It also includes an expanded discussion of JDBC and new coverage of PHP.
　　Chapter 10 collects a number of advanced SQL topics. The discussion of authorization from the old Chapter 8 has been moved here, as has the discussion of recursive SQL from the old Chapter 10. Data cubes, from the old Chapter 20, are now covered here. The rest of the chapter is devoted to the nested-relation model (from the old Chapter 4) and object-relational features of SQL (from the old Chapter 9).
　　Then, Chapters 11 and 12 cover XML and systems based on XML. Except for material at the end of the old Chapter 4, which has been moved to Chapter 11, this material is all new. Chapter 11 covers modeling; it includes expanded coverage of DTD's, along with new material on XML Schema. Chapter 12 is devoted to programming, and it includes sections on XPath, XQuery, and XSLT.
　　Chapter 13 begins the study of database implementation. It covers disk storage and the file structures that are built on disks. This chapter is a condensation of material that, in the first edition, occupied Chapters 11 and 12.
　　Chapter 14 covers index structures, including B-trees, hashing, and structures for multidimensional indexes. This material also condenses two chapters, 13 and 14, from the first edition.
　　Chapters 15 and 16 cover query execution and query optimization, respectively. They are similar to the old chapters of the same numbers. Chapter 17 covers logging, and Chapter 18 covers concurrency control; these chapters are also similar to the old chapters with the same numbers. Chapter 19 contains additional topics on concurrency: recovery, deadlocks, and long transactions. This material is a subset of the old Chapter 19.
　　Chapter 20 is on parallel and distributed databases. In addition to material on parallel query execution from the old Chapter 15 and material on distributed locking and commitment from the old Chapter 19, there are several new sections on distributed query execution: the map-reduce framework for parallel computation, peer-to-peer databases and their implementation of distributed hash tables.
　　Chapter 21 covers information integration. In addition to material on this subject from the old Chapter 20, we have added a section on local-as-view mediators and a section on entity resolution (finding records from several databases that refer to the same entity, e.g., a person).
　　Chapter 22 is on data mining. Although there was some material on the subject in the old Chapter 20, almost all of this chapter is new. It covers association rules and frequent itemset mining, including both the famous A-Priori Algorithm and certain efficiency improvements. Chapter 22 includes the key techniques of shingling, minhashing, and locality-sensitive hashing for finding similar items in massive databases, e.g., Web pages that quote substantially from other Web pages. The chapter concludes with a study of clustering, especially for massive datasets.
　　Chapter 23, all new, addresses two important ways in which the Internet has impacted database technology. First is search engines, where we discuss algorithms for crawling the Web, the well-known PageRank algorithm for evaluating the importance of Web pages, and its extensions. This chapter also covers data-stream-management systems. We discuss the stream data model and SQL language extensions, and conclude with several interesting algorithms for executing queries on streams.
　　Prerequisites ..
　　We have used the book at the "mezzanine" level, in a sequence of courses taken both by undergraduates and by beginning graduate students. The formal prerequisites for the course are Sophomore-level treatments of:
　　1. Data structures, algorithms, and discrete math, and
　　2. Software systems, software engineering, and programming languages.