Automated Floating-Point Precision Analysis
نویسنده
چکیده
Title of dissertation: Automated Floating-Point Precision Analysis Michael O. Lam, Doctor of Philosophy, 2014 Dissertation directed by: Professor Jeffrey K. Hollingsworth Department of Computer Science As scientific computation continues to scale upward, correct and efficient use of floating-point arithmetic is crucially important. Users of floating-point arithmetic encounter many problems, including rounding error, cancellation, and a tradeoff between performance and accuracy. This dissertation addresses these issues by introducing techniques for automated floating-point precision analysis. The contributions include a software framework that enables floating-point program analysis at the binary level, as well as specific techniques for cancellation detection, mixed-precision configuration, and reduced-precision sensitivity analysis. This work demonstrates that automated, practical techniques can provide insights regarding floating-point behavior as well as guidance towards acceptable precision level reduction. The tools and techniques in this dissertation represent novel contributions to the fields of high performance computing and program analysis, and serve as the first major step towards the larger vision of automated floating-point precision and performance tuning. Automated Floating-Point Precision Analysis
منابع مشابه
Profile-Driven Automated Mixed Precision
We present a scheme to automatically set the precision of floating point variables in an application. We design a framework that profiles applications to measure undesirable numerical behavior at the floating point operation level. We use this framework to perform mixed precision analysis to heuristically set the precision of all variables in an application based on their numerical profiles. We...
متن کاملUsing Runtime Floating Point Accuracy Feedback to Make Automated Precision / Performance Improvements or
Using Runtime Floating Point Accuracy Feedback to Make Automated Performance/Precision Improvements or Tradeoffs by Ralph Nathan Department of Electrical and Computer Engineering Duke University Date:_______________________ Approved: ___________________________ Daniel Sorin, Supervisor ___________________________ Andrew Hilton ___________________________ Benjamin Lee ___________________________...
متن کاملFine-grained floating-point precision analysis
Floating-point computation is ubiquitous in high-performance scientific computing, but rounding error can compromise the results of extended calculations, especially at large scales. In this paper, we present new techniques that use binary instrumentation and modification to do fine-grained floating-point precision analysis, simulating any level of precision less than or equal to the precision ...
متن کاملFPGA Based Quadruple Precision Floating Point Arithmetic for Scientific Computations
In this project we explore the capability and flexibility of FPGA solutions in a sense to accelerate scientific computing applications which require very high precision arithmetic, based on IEEE 754 standard 128-bit floating-point number representations. Field Programmable Gate Arrays (FPGA) is increasingly being used to design high end computationally intense microprocessors capable of handlin...
متن کاملSimulation Tools for Fixed Point DSP Algorithms and Architectures
This paper presents software tools that convert the C/C++ floating point source code for a DSP algorithm into a fixed point simulation model that can be used to evaluate the numerical performance of the algorithm on several different fixed point platforms including microprocessors, DSPs and FPGAs. The tools use a novel system for maintaining binary point information so that the conversion from ...
متن کامل