Affiliation:
1. The University of Sydney, Darlington, NSW, Australia
Abstract
Runtime analysis provides an effective method for measuring the sensitivity of programs to rounding errors. To date, implementations have required significant changes to source code, detracting from their widespread application. In this work, we present an open source system that automates the quantitative analysis of floating point rounding errors through the use of C-based source-to-source compilation and a Monte Carlo arithmetic library. We demonstrate its application to the comparison of algorithms, detection of catastrophic cancellation, and determination of whether single precision floating point provides sufficient accuracy for a given application. Methods for obtaining quantifiable measurements of sensitivity to rounding error are also detailed.
Publisher
Association for Computing Machinery (ACM)
Reference73 articles.
1. Design of floating point units for interval arithmetic
2. Estimation of round-off errors on several computer architectures;Asserrhine Jalil;Journal of Universal Computer Science,1995
3. Lecture Notes in Computer Science;Blanchet Bruno
4. Richard P. Brent. 1973. Algorithms for Minimization without Derivatives. Courier Dover Publications. Richard P. Brent. 1973. Algorithms for Minimization without Derivatives. Courier Dover Publications.
5. The design of the Boost interval arithmetic library
Cited by
10 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献