Improvements of BigDFT code in modern HPC architectures

Genovese, Luigi and Videau, Brice and Deutsch, Thierry and Tran, Huan and Stefan Goedecker, and Improvements of BigDFT code in modern HPC architectures. In: UNSPECIFIED.

[img]
Preview
BigDFT.pdf; Text
BigDFT.pdf - Published Version
PRACE Training Material Copyright
The copyright of the structure and formatting of the PRACE Training Portal web site belongs to the Partnership for Advanced Computing AISBL. All training material (including but not limited to course descriptions, presentation files, videos, exercises) are made available for the sole purpose of distance learning by visitors to the Training Portal. If material is labelled with a specific copyright, you should abide to it. Otherwise the copyright of the material belongs to the content creator and the material may not be re-distributed or re-purposed without explicit permission of the content creator.

Download (464kB) | Preview

Abstract

Electronic structure calculations (DFT codes) are certainly among the disciplines for which an increasing of the computational power correspond to an advancement in the scientific results. In this report, we present the ongoing advancements of DFT code that can run on massively parallel, hybrid and heterogeneous CPU-GPU clusters. This DFT code, named BigDFT, is delivered within the GNU-GPL license either in a stand-alone version or integrated in the ABINIT software package. Hybrid BigDFT routines were initially ported with NVidia's CUDA language, and recently more functionalities have been added with new routines writeen within Kronos' OpenCL standard. The formalism of this code is based on Daubechies wavelets, which is a systematic real-space based basis set. The properties of this basis set are well suited for an extension on a GPU-accelerated environment. In addition to focusing on the performances of the MPI and OpenMP parallelisation the BigDFT code, this presentation also relies of the usage of the GPU resources in a complex code with different kinds of operations. A discussion on the interest of present and expected performances of Hybrid architectures computation in the framework of electronic structure calculations is also adressed.

Item Type: Conference or Workshop Item (Lecture)
Subjects: BigDFT
DFT
electronic structure
ABINIT
CUDA
GPU
OpenCL
hybrid
Depositing User: Attila Debreczeni
Date Deposited: 08 Jan 2019 12:49
Last Modified: 08 Jan 2019 12:49
URI: https://materials.prace-ri.eu/id/eprint/134

Actions (login required)

View Item View Item