Berkeley DB - Softcover

Sleepycat Software

Softcover

ISBN 10: 0735710643 ISBN 13: 9780735710641

Publisher: Sams Publishing, 2001

Synopsis

Small, special-purpose computing devices and high-end core Internet servers need fast, reliable database management. Berkeley DB is an embedded database that provides high-performance, scalable, transaction-protected and recoverable data management services to applications. Extremely portable, this library runs under almost all UNIX and Windows variants, as well as a number of embedded, real-time operating systems.

Berkeley DB is the ultimate resource for the world's most widely deployed embedded database engine. This book will aid software architects and engineers, product managers, and systems and network administrators without the overhead imposed by other database products.

Designed by programmers for programmers, this classic library style toolkit provides a broad base of functionality to application writers. This book will help you to make intelligent choices about when and how to use Berkeley DB to meet your needs.

You can visit the Sleepycat website to get the latest errata for this book.

NOTE: The first printing of this book contained an error in the table of contents that caused the page numbers to be off. This will be corrected in the second printing. If you have an earlier edition, you can download a pdf of the correct table of contents that you can print out and use with your book. If you have any questions, please feel free to contact the editor of this book at stephanie.wall@newriders.com.

About this title

About the Author

Sleepycat Software, Inc. develops, and supports Berkeley DB. Berkeley DB was originally written by Keith Bostic, Mike Olson, Margo Seltzer, and Ozan Yigit in the early 1990s. Keith Bostic, one of the principal developers of the University of California's 4.4BSD UNIX releases, approached Seltzer and Yigit to write an Open Source implementation of the UNIX C library dbm hash package for inclusion in 4.4BSD. They agreed, and created a general-purpose hashing library. Later, Bostic enlisted Olson to add a Btree access method to the package. These two projects were integrated and became the first Berkeley DB release, which was included in the 4.4BSD UNIX and related Net/2 releases.

In the years that followed, Berkeley DB was adopted by an enormous number of both proprietary and Open Source projects. Its high performance, ready availability, and easy-to-use interfaces made it a natural tool for a wide variety of applications.

In 1996, there were enough commercial users of the software to justify the formation of a company to enhance and support it. Bostic and Seltzer formed Sleepycat Software to do this. They spent a year adding critical features to the academic code (most notably, transactions, support for concurrent users, and recoverability), and version 2.0 of Berkeley DB was released by Sleepycat Software in 1997.

Since that initial release, Berkeley DB has been deployed in network switches, email clients, wireless communication systems, mission-critical site-monitoring software, and many other applications.

Sleepycat is an Open Source company, and it is committed to the development and distribution of Berkeley DB as an Open Source product. Sleepycat's license for Berkeley DB permits its use at no charge in Open Source applications.

From the Back Cover

Small, special-purpose computing devices and high-end core Internet servers need fast, reliable database management. Berkeley DB is an embedded database that provides high-performance, scalable, transaction-protected and recoverable data management services to applications. Extremely portable, this library runs under almost all UNIX and Windows variants, as well as a number of embedded, real-time operating systems.

Berkeley DB is the ultimate resource for the world's most widely deployed embedded database engine. This book will aid software architects and engineers, product managers, and systems and network administrators without the overhead imposed by other database products.

Designed by programmers for programmers, this classic library style toolkit provides a broad base of functionality to application writers. This book will help you to make intelligent choices about when and how to use Berkeley DB to meet your needs.

You can visit the Sleepycat website to get the latest errata for this book.

NOTE: The first printing of this book contained an error in the table of contents that caused the page numbers to be off. This will be corrected in the second printing. If you have an earlier edition, you can download a pdf of the correct table of contents that you can print out and use with your book. If you have any questions, please feel free to contact the editor of this book at stephanie.wall@newriders.com.

From the Inside Flap

The tool that was missing from our programming hip pockets for the past 20 years was a good data manager. We needed a clean and easy-to-use functional API, transactional support and recoverability, and fantastic performance. It had to scale from single-user applications to servers with thousands of concurrent users. The source code had to be available, so we could debug our applications and integrate it cleanly into many different build environments. When we found that no such tool existed, we wrote our own. The result was Berkeley DB - written by programmers, for programmers. We're proud of Berkeley DB. We hope you like it, too, and that it helps you be a better and more productive programmer.

~Sleepycat Software, June 2001

Excerpt. © Reprinted by permission. All rights reserved.

From the Introduction of this book:

For many programmers, the word "database" instantly calls to mind a complex client/server relational engine from an established commercial vendor. Over the last couple of decades, database management has become big business, and some businesses have become very big indeed, providing products and services for storing information.
With the advent of the Internet, the miniaturization of computing components, and the explosion in data storage capacity of disks and memory systems, computers are being deployed in places and used in ways that were unthinkable just a decade ago. With that shift, the one-size-fits-all philosophy of the big relational products has become outdated. Applications today demand a wide range of data management services: from palmtop devices that need robust, simple single-user record lookup to high-end Internet servers that deliver terabytes of data to thousands of concurrent users. Programmers today need to be able to choose a database manager that delivers exactly the services that their applications demand, and to leave out the complexity and management overhead that their users can't tolerate.

Berkeley DB is an embedded database system. It links directly into the address space of the application that uses it, and provides a simple function-call API for storing and retrieving data. It can run in single-user mode, or provide high-concurrency, transaction-protected, fully recoverable access to records that it stores. Programmers embed it in their applications, and end users are generally unaware that they're using a database at all. There is no separate server to install and administer, no ongoing manual maintenance required, and no complex query language to master.

This book gives detailed information on the design and implementation of Berkeley DB, and how to use it in applications. The book will help you decide whether you need an embedded database at all, and how your application should use Berkeley DB to get the best performance and greatest reliability from it.

What's Inside?
Berkeley DB is organized into two main parts: the Reference Guide and the API Manual. The first section of the Reference Guide has a tutorial, which explains what Berkeley DB is and gives a straightforward example of how it's used in application code. Programmers should be able to read this section and understand how Berkeley DB works generally. The application is a good starting point for developers who need to build applications of their own, and want some code that they can modify.

The Reference Guide gives much more detailed information on the design and implementation of Berkeley DB. This section should help you understand the performance and behavior of your Berkeley DB application. It includes advice on how and when to use particular features, and explains the internals of the library so that you know how it really works.

Part II, the API Manual, provides manual-page style information on the Berkeley DB API. Programmers can use this section to be sure that they're calling Berkeley DB functions with the correct arguments, and to refresh their memories on return types and values and any side effects of individual functions.

Who Is this Book For?
This book is intended for experienced programmers who write applications that need data management services. Applications can be written in C, C++, Java, Perl, Python, Tcl, or PHP; but familiarity with the C programming language will help you get the most out of the book. Berkeley DB is written in C.

You need not be familiar with SQL-based client/server relational systems. If you are, the section that compares Berkeley DB to such systems will help you understand what Berkeley DB is (and is not).

Who Is this Book Not For?
The book is not intended for end users who need a simple GUI-based query tool for their data. Berkeley DB is an embedded database system that requires application code to store and fetch records. Although it's possible to build a GUI-based query tool for a Berkeley DB database, the library, as distributed by Sleepycat Software, does not include that support. The overriding design goal of Berkeley DB is to do one job well: high-performance data management. Extraneous features such as GUI tools, query languages, and relational interfaces are not a part of the core product.

Enjoy!

Edition details

• Publisher: Sams Publishing
• Publication date: 2001
• Language: English
• ISBN 10: 0735710643
• ISBN 13: 9780735710641
• Binding: Paperback
• Edition number: 1
• Number of pages: 688