\texttt{Pselinv} -- a distributed memory parallel algorithm for selected inversion, the symmetric case

DOI10.1145/2786977zbMATH Open1396.65180arXiv1404.0447OpenAlexW2962735411WikidataQ113310236 ScholiaQ113310236MaRDI QIDQ3176312FDOQ3176312

Chao Yang, Lin Lin, Mathias Jacquelin

Publication date: 20 July 2018

Published in: ACM Transactions on Mathematical Software (Search for Journal in Brave)

Abstract: We describe an efficient parallel implementation of the selected inversion algorithm for distributed memory computer systems, which we call exttt{PSelInv}. The exttt{PSelInv} method computes selected elements of a general sparse matrix $A$ that can be decomposed as $A=LU$, where $L$ is lower triangular and $U$ is upper triangular. The implementation described in this paper focuses on the case of sparse symmetric matrices. It contains an interface that is compatible with the distributed memory parallel sparse direct factorization exttt{SuperLU_DIST}. However, the underlying data structure and design of exttt{PSelInv} allows it to be easily combined with other factorization routines such as exttt{PARDISO}. We discuss general parallelization strategies such as data and task distribution schemes. In particular, we describe how to exploit the concurrency exposed by the elimination tree associated with the $LU$ factorization of $A$. We demonstrate the efficiency and accuracy of exttt{PSelInv} by presenting a number of numerical experiments. In particular, we show that exttt{PSelInv} can run efficiently on more than $4,000$ cores for a modestly sized matrix. We also demonstrate how exttt{PSelInv} can be used to accelerate large-scale electronic structure calculations.

Full work available at URL: https://arxiv.org/abs/1404.0447

Recommendations

• SelInv---An Algorithm for Selected Inversion of a Sparse Symmetric Matrix
• A Fast Parallel Algorithm for Selected Inversion of Structured Sparse Matrices with Application to 2D Electronic Structure Calculations
• Fast sparse selected inversion
• A note on parallel matrix inversion
• Parallel computation of entries of \({A}^{-1}\)

zbMATH Keywords

electronic structure theoryselected inversionhigh-performance computationsparse direct methoddistributed memory parallel algorithm

Mathematics Subject Classification ID

Direct numerical methods for linear systems and matrix inversion (65F05) Parallel numerical computation (65Y05)

Cited In (5)

• Efficient Covariance Approximations for Large Sparse Precision Matrices
• Numerical solution of large scale Hartree–Fock–Bogoliubov equations
• SelInv---An Algorithm for Selected Inversion of a Sparse Symmetric Matrix
• Pselinv
• Improvements on non-equilibrium and transport Green function techniques: the next-generation Transiesta

Uses Software

• SuperLU-DIST
• ScaLAPACK
• SparseMatrix
• PT-Scotch
• Harwell-Boeing sparse matrix collection

This page was built for publication: \texttt{Pselinv} -- a distributed memory parallel algorithm for selected inversion, the symmetric case

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3176312)