A review on Approximate Computing applied to meteorological forecast models using software-based techniques

• Marcelo Augusto Sudo UNIFESP http://orcid.org/0000-0001-9497-3740
• Álvaro Luiz Fazenda UNIFESP https://orcid.org/0000-0002-4052-1113

DOI: https://doi.org/10.5753/eradsp.2020.16884

Abstract

This study is part of research on Approximate Computing software-based techniques applied to meteorological forecast numerical models. Currently, a Systematic Review is ongoing, and it will serve as a reference to verify which algorithms have already been ported and which techniques are being used. This document focuses only on some approximated computing techniques, especially on mixed or reduced precision. Conclusions up to now show just a few research articles regarding the subject, so there are still many aspects to unveil.

Keywords: Approximate Computing, mixed precision, weather forecasting

References

Lim, Y. and Tillmann, U. (1986) “End-to-End Speech Freeze-Out Fractions in a Network with Speech Interpolations”, IEEE Journals , vol. 34, issue 12, pages 1236–1245.

Han, J. and Orshansky, M. (2013) “Approximate computing: An emerging paradigm for energy-efficient design”, IEEE Conferences, pages 1–6.

Li, C., Sengupta, D., Snigdha, F., Xu, W., Hu, J. and Sapatnekar, S. (2017) “Special session: a quantifiable approach to approximate computing”, IEEE Conferences, pages 1–2.