Dynamic floating-point cancellation detection
نویسندگان
چکیده
منابع مشابه
Dynamic floating-point cancellation detection
As scientific computation continues to scale, it is crucial to use floating-point arithmetic processors as efficiently as possible. Lower precision allows streaming architectures to perform more operations per second and can reduce memory bandwidth pressure on all architectures. However, using a precision that is too low for a given algorithm and data set will result in inaccurate results. Thus...
متن کاملPractical Floating-Point Divergence Detection
Reducing floating-point precision allocation in HPC programs is of considerable interest from the point of view of obtaining higher performance. However, this can lead to unexpected behavioral deviations from the programmer’s intent. In this paper, we focus on the problem of divergence detection: when a given floating-point program exhibits different control flow (or differs in terms of other d...
متن کاملA Dynamic Analysis for Tuning Floating-point Precision
Floating-point numbers are widely used to approximate real number arithmetic in applications from domains such as scientific computing, graphics, and finance. However, squeezing one or more real numbers into a finite number of bits requires an approximate representation. As a consequence, the result of a floating-point computation typically contains numerical errors. To minimize the chance of p...
متن کاملEstimation of Cancellation Errors in Multivariate Hensel Construction with Floating-Point Numbers
Multivariate Hensel construction with floating-point numbers often cause large cancellation errors which are errors due to a cancellation of almost the same numbers. Sasaki and Yamaguchi [SY98] showed that multivariate Hensel construction causes large cancellation errors if the expansion point is chosen near a singular point, and Sasaki [Sas00] studied four mechanisms of term cancellations near...
متن کاملCustomising Floating-Point Designs
This paper describes a method for customising the representation of floating-point numbers that exploits the flexibility of reconfigurable hardware. The method determines the appropriate size of mantissa and exponent for each operation in a design, so that a cost function with a given error specification for the output relative to a reference representation can be satisfied. Currently our tool,...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Parallel Computing
سال: 2013
ISSN: 0167-8191
DOI: 10.1016/j.parco.2012.08.002