NekRS, a GPU-accelerated spectral element Navier–Stokes solver

Download
2022-12-01
Fischer, Paul
Kerkemeier, Stefan
Min, Misun
Lan, Yu-Hsiang
Phillips, Malachi
Rathnayake, Thilina
Merzari, Elia
Tomboulides, Ananias
Karakuş, Ali
Chalmers, Noel
Warburton, Tim
The Author(s)The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B grid points (240B degrees-of-freedom).
Parallel Computing

Suggestions

Analysis of the energy justice in natural gas distribution with Multiscale Geographically Weighted Regression (MGWR)
Şenyel Kürkçüoğlu, Müzeyyen Anıl (2023-12-01)
The Author(s)Energy justice is violated when particular customers and locations are excluded from a variety of urban energy service distribution. This study explores energy justice in terms of natural gas distribution by providing empirical evidence from 356 neighborhoods of Izmir Metropolitan Area (IMA). The aim is to reveal driving factors of natural gas investment and the spatial reflections of the relationships between investment, and socio-economic and physical characteristics. A global regression mode...
Determination of Nonlinear Joint Forces and Nonlinear Identification of Jointed Connections Using FRFs
Soleimani, Hossein; Ciğeroğlu, Ender; Özgüven, Hasan Nevzat (2023-01-01)
The Author(s), under exclusive license to Springer Nature Switzerland AG.Identification and modeling of joint dynamics is an important issue in the dynamic modeling of structures with joints. Moreover, structures connected by joints may show nonlinear behavior due to the existence of dry friction between the connected elements, especially at higher levels of vibration. Therefore, nonlinear identification of joints and parameterization of them is a necessary step in the dynamic analysis of structures. In thi...
Compatibility of the dimensional reduction and variation procedures for a quadratic curvature model with a Kaluza–Klein Ansatz
Başkal, Sibel; Çelik, Sinan (2022-10-01)
The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.The introduction of extra dimensions is an invaluable strategy for the unification of gravity with other physical fields. Nevertheless, the matter in hand is to be eventually reduced to the actual 4D spacetime. The Kaluza–Klein theory is no exception to this well-known scheme. There are two procedures to obtain the field equations from a higher dimensional action. One can either take the variation of the ...
Collision-Free Motion Planning for an Aligned Multiple-Turret System Operating in Extreme Environment
Yerlikaya, Ümit; Balkan, Raif Tuna (2021-01-01)
© The Author(s), 2021. Published by Cambridge University Press.Instead of using the tedious process of manual positioning, an off-line path planning algorithm has been developed for military turrets to improve their accuracy and efficiency. In the scope of this research, an algorithm is proposed to search a path in three different types of configuration spaces which are rectangular-, circular-, and torus-shaped by providing three converging options named as fast, medium, and optimum depending on the applica...
Efficient exascale discretizations: High-order finite element methods
Kolev, Tzanio; et. al. (2021-01-01)
© The Author(s) 2021.Efficient exploitation of exascale architectures requires rethinking of the numerical algorithms used in many large-scale applications. These architectures favor algorithms that expose ultra fine-grain parallelism and maximize the ratio of floating point operations to energy intensive data movement. One of the few viable approaches to achieve high efficiency in the area of PDE discretizations on unstructured grids is to use matrix-free/partially assembled high-order finite element metho...
Citation Formats
P. Fischer et al., “NekRS, a GPU-accelerated spectral element Navier–Stokes solver,” Parallel Computing, vol. 114, pp. 0–0, 2022, Accessed: 00, 2023. [Online]. Available: https://hdl.handle.net/11511/101817.