rpe v5: an emulator for reduced floating-point precision in large numerical simulations
Open Access
- 16 June 2017
- journal article
- research article
- Published by Copernicus GmbH in Geoscientific Model Development
- Vol. 10 (6) , 2221-2230
Abstract
This paper describes the rpe (reduced-precision emulator) library which has the capability to emulate the use of arbitrary reduced floating-point precision within large numerical models written in Fortran. The rpe software allows model developers to test how reduced floating-point precision affects the result of their simulations without having to make extensive code changes or port the model onto specialized hardware. The software can be used to identify parts of a program that are problematic for numerical precision and to guide changes to the program to allow a stronger reduction in precision.The development of rpe was motivated by the strong demand for more computing power. If numerical precision can be reduced for an application under consideration while still achieving results of acceptable quality, computational cost can be reduced, since a reduction in numerical precision may allow an increase in performance or a reduction in power consumption. For simulations with weather and climate models, savings due to a reduction in precision could be reinvested to allow model simulations at higher spatial resolution or complexity, or to increase the number of ensemble members to improve predictions. rpe was developed with a particular focus on the community of weather and climate modelling, but the software could be used with numerical simulations from other domains.Keywords
All Related Versions
Funding Information
- European Research Council (291406)
This publication has 25 references indexed in Scilit:
- The use of imprecise processing to improve accuracy in weather & climate predictionJournal of Computational Physics, 2014
- On the use of inexact, pruned hardware in atmospheric modellingPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2014
- Stochastic Electronics: A Neuro-Inspired Design Paradigm for Integrated CircuitsProceedings of the IEEE, 2014
- Rossby rip currentsGeophysical Research Letters, 2013
- Stochastic Arithmetic in MultiprecisionMathematics in Computer Science, 2011
- Toward a New Generation of World Climate Research and Computing FacilitiesBulletin of the American Meteorological Society, 2010
- Accuracy and Performance of Single versus Double Precision Arithmetics for Maximum Likelihood Phylogeny Reconstruction2010
- CADNA: a library for estimating round-off error propagationComputer Physics Communications, 2008
- Numerical ‘health check’ for scientific codes: the CADNA approachComputer Physics Communications, 2007
- Computational Design of the Basic Dynamical Processes of the UCLA General Circulation Model1977