並列分子科学計算プログラムの電力制約下における性能最適化 [Published online J. Comput. Chem. Jpn., 14, 201-202, by J-STAGE]

[Published online Journal of Computer Chemistry, Japan Vol.14, 201-202, by J-STAGE]
<Title:> 並列分子科学計算プログラムの電力制約下における性能最適化
<Author(s):> 稲富 雄一, 井上 弘士
<Corresponding author E-Mill:> yuichi.inadomi(at)cpc.ait.kyushu-u.ac.jp
<Abstract:> Power is the most important resource on the next-generation supercomputers, and they will be operated under power constraint. Therefore, there is a need to maximize performance of HPC application under power constraint. To do such optimization, we’ve developed and reported a method to improve performance by power allocation for each processor, which is called the variation-aware power budgeting. In this study, we carried out large-scale performance evaluation of a proposed method for two mini-applications related to molecular science, Modylas-mini and NTChem-mini. As a result, our method can improve their performance under power constraint up to 1.99 times speedup compared to conventional power constraint.
<Keywords:> Power-constrained supercomputing, Variation-aware power budgeting
<URL:> https://www.jstage.jst.go.jp/article/jccj/14/6/14_2015-0066/_article/-char/ja/