Optimizing Precision for High-Performance, Robust, and Energy-Efficient Computations

Roman Iakymchuk (1) , Stef Graillat (1) , Fabienne Jézéquel (2, 1) , Daichi Mukunoki , Toshiyuki Imamura , Yiyu Tan , Atsushi Koshiba , Jens Huthmann , Kentaro Sano , Norihisa Fujita , Taisuke Boku (3)

1 PEQUAN - Performance et Qualité des Algorithmes Numériques
2 UP2 - Université Panthéon-Assas
3 Graduate School of Systems and Information Engineering [Tsukuba]

In numerical computations, precision of floating-point computations is a key factor to determine the performance (speed and energy-efficiency) as well as the reliability (accuracy and reproducibility). However, precision generally plays a contrary role for both. Therefore, the ultimate concept for maximizing both at the same time is the minimal-precision computing through precision-tuning, which adjusts the optimal precision for each operation and data. Several studies have been already conducted for it so far, but the scope of those studies is limited to the precision-tuning alone. Our project aims to propose a broader concept of the minimal-precision computing system with precision-tuning, involving both hardware and software stack. This approach is robust, general, comprehensive, high-performant, and realistic.

Domaines: Arithmétique des ordinateurs; Calcul parallèle, distribué et partagé [cs.DC]; Performance et fiabilité [cs.PF]

Roman Iakymchuk, Stef Graillat, Fabienne Jézéquel, Daichi Mukunoki, Toshiyuki Imamura, et al.. Optimizing Precision for High-Performance, Robust, and Energy-Efficient Computations. International Conference on High Performance Computing in Asia-Pacific Region,, Jan 2020, Fukuoka, Japan. ⟨hal-02401813⟩