Most numerical data in commercial and human-centric applications are decimal, and decimal floating-point arithmetic has become increasingly important in recent years as these applications become ever more complex.
Recent benchmarking has shown that many Java applications spend as much as 50% to 90% of their time in the BigDecimal floating-point operations, so the BigDecimal class has had major enhancements in Java 5.0. It is now faster and much easier to use. In addition, IBM is now implementing decimal floating-point hardware; the new BigDecimal can use this hardware without need for recompilation, which will give dramatic performance improvements.
In this session, Mike will cover:
- why and when you need to use BigDecimal arithmetic in your applications
- why IEEE 754, XML Schema, and other standards groups are adding decimal floating-point to their specifications
- how to use the new java.math decimal classes for correct results and best performance.
Mike Cowlishaw has worked in both hardware and software design and is the leader of the IBM decimal arithmetic initiative and author of the Java 5.0 BigDecimal changes. He introduced Java to the Colorado Software Summit, and has long been interested in the human aspects of computing, including the Rexx programming language, color perception, text processing, and decimal arithmetic. He is an IBM Fellow, a Fellow of the Royal Academy of Engineering, and a Visiting Professor in the Department of Computer Science at the University of Warwick, UK.