Java Performance and Scalability: 1 (The Sun Microsystems Press Java series) (Paperback)

The 48 optimisations for various aspects of the Java API make up the heart of this text. Early sections look at a particularly tricky subject--string processing--giving you several techniques for increasing speed and reducing object creation. The in-depth examination of the costs and benefits of string concatenation and the immutability of Java String objects will help you uncover possible hot spots that cause lagging performance with strings.

Subsequent sections look at more general strategies for faster programs, like purging obsolete code and the well-known 80-20 rule. (Optimising the 20 percent of your code that consumes 80 percent of processing time, allowing you to create faster software. This book shows you how.) Short tips on such APIs as Java collections and the Java Native Interface (JNI) come next.

The second half of the book zeroes in on better server-side performance with tips on caching and file I/O. The guide to multithreading and synchronisation is a standout here. You'll get several smart tips for speeding up servlets (like pre-computing your standard strings for headers and other boilerplate HTML). The book closes with a valuable custom class for JDBC database connection pooling and a custom HTTP server. Both samples make use of the techniques presented earlier in the text. Even if you don't adopt every suggestion, the ideas here can get you thinking about performance in new ways. (One of the main points of this book is that you can write faster custom code instead of always relying on built-in Java classes and APIs.)

Clearly, Java performance is not something that happens by accident. It must be engineered into your code. This compilation of useful techniques and tips deserves a close look by anyone who wants to squeeze more performance out of Java. --Richard Dragan, amazon.com

Topics covered:

• Introduction to Java performance
• More efficient Java string handling (optimising String, String Buffer, and String Tokenizer objects)
• Reducing Java overhead
• Faster initialisation
• The 80-20 rule for optimisation
• Getting rid of obsolete code
• Speeding up the JDK 1.1 Vector and Hashtable collection classes
• Caching techniques
• Optimising file I/O and buffering
• Object recycling and object pools
• Techniques for multithreading and synchronisation
• Amdahl's law and scalability basics
• Pros and cons of using the Java Native Interface (JNI)
• Hints for optimising Remote Invocation Method (RMI) calls
• Performance hints for servlets and JavaServer Pages (JSPs)
• Custom sample code for JDBC database connection pooling and a Java HTTP server.

Performance is now the area where Java is being put to the test, offering ever greater competitive advantage for those who master it. In this book, one of the world's leading software optimization experts presents 48 lessons that address virtually every aspect of Java server-side performance, Java Performance and Scalability, Volume I proves that, through careful attention to underlying source code and program design, the Java programming language can meet and exceed developers' greatest expectations. Dov Bulka targets the most common and critical Java performance pitfalls, offering dozens of practical tips and solutions covering each performance-critical area. Bulka has done the hard work of digging deep into the JDK to understand Java performance in unprecedented depth. This book includes not only code examples, but actual measurements quantifying the improvements they deliver. Coverage includes strings, containers, numerics, I/O, RMI, JDBC, servlets, JSP, and much more -- including a complete case study that walks through improving the performance of a Web server by a full order of magnitude.