Filters








60 Hits in 2.0 sec

Optimizations of the Eigensolvers in the ELPA Library [article]

P. Kus, A. Marek, S. S. Koecher, H.-H. Kowalski, C. Carbogno, Ch. Scheurer, K. Reuter, M. Scheffler, H. Lederer
2018 arXiv   pre-print
and of performance-critical parameters used in the ELPA-library.  ...  On some examples of scientific interest, simulated with the FHI-aims application, the advantages of the latest optimizations of the ELPA-library are demonstrated.  ...  Acknowledgments Part of this work is co-funded by BMBF grant 01IH15001 of the German Government.  ... 
arXiv:1811.01277v1 fatcat:g6hy6fifznfefi3bjr3xadex2y

Numerical Library Eigensolver Performance On Prace Tier-0 Systems

Andrew Sunderland
2017 Zenodo  
This version of the report is updated with results from ScaLAPACK and with the recently released ELPA library routines on the PRACE Tier-0 machines CURIE and JUGENE.  ...  Parallel eigensolver operations are at the computational core of many large-scale scientific and engineering application codes.  ...  The author would also like to express his gratitude to Ian Bush of NAG Ltd and Hermann Lederer of RZG in Germany.  ... 
doi:10.5281/zenodo.810719 fatcat:5d525xjwzbe2fnnhilcwal762q

ELSI – An Open Infrastructure for Electronic Structure Solvers [article]

Victor Wen-zhe Yu, Carmen Campos, William Dawson, Alberto García, Ville Havu, Ben Hourahine, William P Huhn, Mathias Jacquelin, Weile Jia, Murat Keçeli, Raul Laasner, Yingzhou Li (+7 others)
2020 arXiv   pre-print
The ELectronic Structure Infrastructure (ELSI) project provides an open-source software interface to facilitate the implementation and optimal use of high-performance solver libraries covering cubic scaling  ...  eigensolvers, linear scaling density-matrix-based algorithms, and other reduced scaling methods in between.  ...  pentadiagonalization-based dense eigensolver in the EigenExa library, and the GPU-accelerated, shared-memory dense eigensolvers in the MAGMA library. 2.  ... 
arXiv:1912.13403v2 fatcat:32x5twwzgrhsvhtufjcf2rb4ke

A hybrid Hermitian general eigenvalue solver [article]

Raffaele Solcà, Thomas C. Schulthess, Azzam Haidar, Stanimire Tomov, Ichitaro Yamazaki, Jack Dongarra
2012 arXiv   pre-print
The resulting eigensolvers are state-of-the-art in HPC, significantly outperforming existing libraries.  ...  The adoption of hybrid GPU-CPU nodes in traditional supercomputing platforms opens acceleration opportunities for electronic structure calculations in materials science and chemistry applications, where  ...  The authors would also like to thank the National Science Foundation, the Department of Energy, NVIDIA, and the MathWorks for this research effort.  ... 
arXiv:1207.1773v1 fatcat:2m7iyy4fhjf6rlipewwiartbnm

Benefits from using mixed precision computations in the ELPA-AEO and ESSEX-II eigensolver projects [article]

Andreas Alvermann, Achim Basermann, Hans-Joachim Bungartz, Christian Carbogno, Dominik Ernst, Holger Fehske, Yasunori Futamura, Martin Galgon, Georg Hager, Sarah Huber, Thomas Huckle, Akihiro Ida, Akira Imakura, Masatoshi Kawai (+16 others)
2018 arXiv   pre-print
We first briefly report on the status and recent achievements of the ELPA-AEO (Eigenvalue Solvers for Petaflop Applications - Algorithmic Extensions and Optimizations) and ESSEX II (Equipping Sparse Solvers  ...  Then we focus on a topic addressed in both projects, the use of mixed precision computations to enhance efficiency.  ...  Such methods are available in the widely used ELPA library [19] , which had originated in an earlier project [2] and is being improved further and extended with ELPA-AEO.  ... 
arXiv:1806.01036v1 fatcat:is32tm7p6zba5ppt5hugufv32q

GPU-Acceleration of the ELPA2 Distributed Eigensolver for Dense Symmetric and Hermitian Eigenproblems [article]

Victor Wen-zhe Yu, Jonathan Moussa, Pavel Kůs, Andreas Marek, Peter Messmer, Mina Yoon, Hermann Lederer, Volker Blum
2021 arXiv   pre-print
We here present GPU-oriented optimizations of the ELPA two-stage tridiagonalization eigensolver (ELPA2).  ...  Finally, we demonstrate the performance of the GPU-accelerated eigensolver developed in this work for routine semi-local KS-DFT calculations comprising thousands of atoms.  ...  Conclusions In this paper, we report our GPU-oriented optimizations of the two-stage tridiagonalization eigensolver ELPA2 in the ELPA library.  ... 
arXiv:2002.10991v4 fatcat:khp5rdi7lvhfzn2nesknyo7gju

Leading Edge Hybrid Multi-GPU Algorithms for Generalized Eigenproblems in Electronic Structure Calculations [chapter]

Azzam Haidar, Raffaele Solcà, Mark Gates, Stanimire Tomov, Thomas Schulthess, Jack Dongarra
2013 Lecture Notes in Computer Science  
We describe and analyze a successful methodology to address the challenges-starting from our algorithm design, kernel optimization and tuning, to our programming model-in the development of a scalable  ...  We report the performance impact on the generalized eigensolver when different fractions of eigenvectors are needed.  ...  The authors would like to thank the National Science Foundation, the Department of Energy, NVIDIA, and MathWorks for supporting this research effort.  ... 
doi:10.1007/978-3-642-38750-0_6 fatcat:xd4jb5ccxzbfhpr4zipd5sohpy

The ELPA library: scalable parallel eigenvalue solutions for electronic structure theory and computational science

A Marek, V Blum, R Johanni, V Havu, B Lang, T Auckenthaler, A Heinecke, H-J Bungartz, H Lederer
2014 Journal of Physics: Condensed Matter  
ELPA addresses standard as well as generalized eigenvalue problems, relying on the well documented matrix layout of the Scalable Linear Algebra PACKage (ScaLAPACK) library but replacing all actual parallel  ...  For these steps, ELPA significantly outperforms the corresponding ScaLAPACK routines and proprietary libraries that implement the ScaLAPACK interface (e.g. Intel's MKL).  ...  Acknowledgments The ELPA library was developed by a multidisciplinary consortium of scientists from the Fritz Haber Institute of the Max Planck Society, the Garching Computing Center of the Max Planck  ... 
doi:10.1088/0953-8984/26/21/213201 pmid:24786764 fatcat:kagmuswa4nd43ioehd3ughhid4

A novel hybrid CPU–GPU generalized eigensolver for electronic structure calculations based on fine-grained memory aware tasks

Azzam Haidar, Stanimire Tomov, Jack Dongarra, Raffaele Solcà, Thomas Schulthess
2013 The international journal of high performance computing applications  
The resulting eigensolvers are state-ofthe-art in HPC, significantly outperforming existing libraries.  ...  The adoption of hybrid GPU-CPU nodes in traditional supercomputing platforms such as the Cray-XK6 opens acceleration opportunities for electronic structure calculations in materials science and chemistry  ...  ACKNOWLEDGMENTS The authors would like to thank the National Science Foundation, the Department of Energy, NVIDIA, and MathWorks  ... 
doi:10.1177/1094342013502097 fatcat:ytkh7zbpfzhjrms4xif3vqso2u

NSF SI2-SSI ELSI

Volker Blum
2017 Figshare  
In addition to work on the interface itself, the project also focuses on further developing solution strategies such as the ELPA eigensolver library, the PEXSI ON(N^2) scaling approach, the orbital minimization  ...  Abstract:The ELSI ("Electronic Structure Infrastructure") project seeks to develop an interface to several software libraries created to solve or circumvent the algebraic eigenvalue problem of Kohn-Sham  ...  Any opinions, findings, and conclusions or recommendations expressed here are those of the authors and do not necessarily reflect the views of the National Science Foundation.  ... 
doi:10.6084/m9.figshare.4569544.v2 fatcat:yl465k4wj5anhdqvtwxsykl7a4

NSF SI2-SSI ELSI

Volker Blum
2017 Figshare  
In addition to work on the interface itself, the project also focuses on further developing solution strategies such as the ELPA eigensolver library, the PEXSI ON(N^2) scaling approach, the orbital minimization  ...  Abstract:The ELSI ("Electronic Structure Infrastructure") project seeks to develop an interface to several software libraries created to solve or circumvent the algebraic eigenvalue problem of Kohn-Sham  ...  Any opinions, findings, and conclusions or recommendations expressed here are those of the authors and do not necessarily reflect the views of the National Science Foundation.  ... 
doi:10.6084/m9.figshare.4569544.v3 fatcat:bhjoxdym3rfilawx7hgafueztm

NSF SI2-SSI ELSI

Volker Blum
2017 Figshare  
In addition to work on the interface itself, the project also focuses on further developing solution strategies such as the ELPA eigensolver library, the PEXSI ON(N^2) scaling approach, the orbital minimization  ...  Abstract:The ELSI ("Electronic Structure Infrastructure") project seeks to develop an interface to several software libraries created to solve or circumvent the algebraic eigenvalue problem of Kohn-Sham  ...  Any opinions, findings, and conclusions or recommendations expressed here are those of the authors and do not necessarily reflect the views of the National Science Foundation.  ... 
doi:10.6084/m9.figshare.4569544.v1 fatcat:c7vm2ngwh5f4bjoqt4lxfm4ozq

Abstract: A Novel Hybrid CPU-GPU Generalized Eigensolver for Electronic Structure Calculations Based on Fine Grained Memory Aware Tasks

Raffaele Solca, Azzam Haidar, Stanimire Tomov, Thomas C. Schulthess, Jack Dongarra
2012 2012 SC Companion: High Performance Computing, Networking Storage and Analysis  
The resulting eigensolvers are state-of-the-art in high-performance computing, significantly outperforming existing libraries.  ...  The adoption of hybrid CPU-GPU nodes in traditional supercomputing platforms such as the Cray-XK6 opens acceleration opportunities for electronic structure calculations in materials science and chemistry  ...  The comparison shows that we overcome in time to solution the shared-memory MKL library and the distributed-memory ELPA library.  ... 
doi:10.1109/sc.companion.2012.173 dblp:conf/sc/SolcaHTSD12 fatcat:jy4vxkbgjvhb3n33mqvvnwpy6a

ChASE – A Distributed Hybrid CPU-GPU Eigensolver for Large-scale Hermitian Eigenvalue Problems [article]

Xinzhe Wu and Davor Davidovic and Sebastian Achilles and Edoardo Di Napoli
2022 arXiv   pre-print
The Chebyshev Accelerated Subspace Eigensolver (ChASE) is a modern subspace eigensolver to compute partial extremal eigenpairs of large-scale Hermitian eigenproblems with the acceleration of a filter based  ...  Recently, there have been significant scientific contributions to the development of filter-based subspace eigensolver to compute partial eigenspectrum.  ...  ACKNOWLEDGMENTS This work was supported by the Croatian Science Foundation under grant number HRZZ-UIP-2020-02-4559, by the Ministry of Science and Education of the Republic of Croatia and the Deutsche  ... 
arXiv:2205.02491v1 fatcat:zzuld3ztojektogendmj5iiuda

ELSI: A unified software interface for Kohn–Sham electronic structure solvers

Victor Wen-zhe Yu, Fabiano Corsetti, Alberto García, William P. Huhn, Mathias Jacquelin, Weile Jia, Björn Lange, Lin Lin, Jianfeng Lu, Wenhui Mi, Ali Seifitokaldani, Álvaro Vázquez-Mayagoitia (+3 others)
2018 Computer Physics Communications  
Currently supported algorithms include the dense generalized eigensolver library ELPA, the orbital minimization method implemented in libOMM, and the pole expansion and selected inversion (PEXSI) approach  ...  The ELSI interface aims to simplify the implementation and optimal use of the different strategies, by offering (a) a unified software framework designed for the electronic structure solvers in Kohn-Sham  ...  eigenproblem using the ELPA eigensolver library.  ... 
doi:10.1016/j.cpc.2017.09.007 fatcat:mavgakxxyrfvhetlypj4ktzmri
« Previous Showing results 1 — 15 out of 60 results