Performance Optimization and Evaluation of Scalable Optoelectronics Application on Large Scale KNL Cluster

Hirokawa, Yuta; Boku, Taisuke; Uemoto, Mitsuharu; Sato, Shunsuke A.; Yabana, Kazuhiro

doi:10.1007/978-3-319-92040-5_11

Cited by 1 publication

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We next consider the hand-coding vectorization using ACLE. In Hirokawa et al (2018), we discussed optimization of the same calculation using Intel AVX-512 intrinsics. In SVE (Scalable Vector Extension) instructions that A64FX supports, it is possible to set the hardware SIMD length between 128-bit and 2048- bit.…”

Section: Simulation Methods and Implementationsmentioning

confidence: 99%

See 1 more Smart Citation

Large-scale ab initio simulation of light–matter interaction at the atomic scale in Fugaku

Hirokawa

Yamada

et al. 2022

The International Journal of High Performance Computing Applica

Self Cite

View full text Add to dashboard Cite

In the field of optical science, it is becoming increasingly important to observe and manipulate matter at the atomic scale using ultrashort pulsed light. For the first time, we have performed the ab initio simulation solving the Maxwell equation for light electromagnetic fields, the time-dependent Kohn-Sham equation for electrons, and the Newton equation for ions in extended systems. In the simulation, the most time-consuming parts were stencil and nonlocal pseudopotential operations on the electron orbitals as well as fast Fourier transforms for the electron density. Code optimization was thoroughly performed on the Fujitsu A64FX processor to achieve the highest performance. A simulation of amorphous SiO2 thin film composed of more than 10,000 atoms was performed using 27,648 nodes of the Fugaku supercomputer. The simulation achieved excellent time-to-solution with the performance close to the maximum possible value in view of the memory bandwidth bound, as well as excellent weak scalability.

show abstract

Section: Simulation Methods and Implementationsmentioning

confidence: 99%

“…A typical single production run required 19 hours using 8000 nodes (64,000 cores). In Hirokawa et al (2018), coupled calculation of 3D Maxwell and 3D TDKS using ARTED was reported. The calculation was performed using almost all 8192 nodes of the KNL processors of Oakforest-PACS at JCAHPC, and a 12–16% performance was obtained.…”

Section: Introductionmentioning

confidence: 99%