Physical Database Design

the database professional's guide to exploiting indexes, views, storage, and more


  • Sam Lightstone, IBM, Toronto, Canada
  • Toby Teorey, University of Michigan, Ann Arbor, USA
  • Tom Nadeau, Ubiquiti Inc., Ann Arbor, MI

The rapidly increasing volume of information contained in relational databases places a strain on databases, performance, and maintainability: DBAs are under greater pressure than ever to optimize database structure for system performance and administration. Physical Database Design discusses the concept of how physical structures of databases affect performance, including specific examples, guidelines, and best and worst practices for a variety of DBMSs and configurations. Something as simple as improving the table index design has a profound impact on performance. Every form of relational database, such as Online Transaction Processing (OLTP), Enterprise Resource Management (ERP), Data Mining (DM), or Management Resource Planning (MRP), can be improved using the methods provided in the book.
View full description


Database designers and database administrators working with large databases and data warehouses, who want to improve processing time for transactions and information access, resolve storage issues, and make life better for all of us.


Book information

  • Published: March 2007
  • ISBN: 978-0-12-369389-1


I highly recommend Physical Database Design by Lightstone, Teorey, and Nadeau. The book covers fine aspects of physical design -- issues such as the effects of different approaches to indexes, tradeoffs in materializing views, and details of physical data layout. Unlike other books, it does not focus on a particular product, but instead covers the deep principles that cut across products. The book addresses both transaction intensive applications (OLTP) as well as data warehouses (OLAP). Their new book is a welcome addition to the literature. --Michael Blaha, OMT Associates, Inc. This is an excellent book on physical database design, giving pragmatic models and advice. It has a wealth of information for both the student and for the practitioner -- presenting analytic models and practical tips that are demonstrated with examples using Oracle, DB2, and Microsoft SQL Server. --Jim Gray, Microsoft Research

Table of Contents

1 Introduction to Physical Database Design2 Basic Indexing Methods3 Query Optimization and Plan Selection4 Selecting Indexes5 Selecting Materialized Views6 Shared-nothing Partitioning7 Range Partitioning8 Multidimensional Clustering9 The Interdependence Problem10 Counting and Data Sampling in Physical Design Exploration11 Query Execution Plans and Physical Design12 Automated Physical Database Design13 Down to the Metal: Server Resources and Topology14 Physical Design for Decision Support, Warehousing, and OLAP15 Denormalization16 Distributed Data AllocationAppendix A A Simple Performance Model for DatabasesAppendix B Technical Comparison of DB2 HADR with Oracle Data Guard for Database Disaster Recovery