Paquet : libsuperlu-dist-dev (8.2.1+dfsg1-5 et autres)

[ Paquet source : superlu-dist  ]

Highly distributed solution of sparse linear equations

SuperLU is a general purpose library for the direct solution of large, sparse, nonsymmetric systems of linear equations. The library is written in C and is callable from either C or Fortran program. It uses MPI, OpenMP and CUDA to support various forms of parallelism. It supports both real and complex datatypes, both single and double precision, and 64-bit integer indexing. The library routines performs an LU decomposition with partial pivoting and triangular system solves through forward and back substitution. The LU factorization routines can handle non-square matrices but the triangular solves are performed only for square matrices. The matrix columns may be preordered (before factorization) either through library or user supplied routines. This preordering for sparsity is completely separate from the factorization. Working precision iterative refinement subroutines are provided for improved backward stability. Routines are also provided to equilibrate the system, estimate the condition number, calculate the relative backward error, and estimate error bounds for the refined solutions.

SuperLU_DIST implements the algorithms for distributed memory, targetting highly parallel distributed memory hybrid systems. The numerical factorization routines are already implemented for hybrid systems with multiple GPUs. Further work will be needed to implement the other phases of the algorithms on the hybrid systems and to enhance strong scaling to extreme scale.

The main library is libsuperlu_dist.so but a fortran wrapper library is also provided as libsuperlu_dist_fortran.so

This package provides development files for building client applications against superlu-dist.

Étiquettes: Développement de logiciel: Bibliothèques, Rôle: Bibliothèque de programmation

Autres paquets associés à libsuperlu-dist-dev

• dep: gfortran-14 [amd64, armel, armhf, hppa, i386, m68k, sh4, x32]

  compilateur Fortran GNU

  ou gfortran-mod-15

  paquet virtuel fourni par flang-15, flang-16, flang-17, flang-18, flang-19, gfortran, gfortran-10, gfortran-11, gfortran-12, gfortran-13, gfortran-14, gfortran-9

• dep: libblas3

  implémentation de référence en algèbre linéaire de base - bibliothèque partagée

  ou libblas.so.3

  paquet virtuel fourni par libatlas3-base, libblas3, libblis4-openmp, libblis4-pthread, libblis4-serial, libopenblas0-openmp, libopenblas0-pthread, libopenblas0-serial

• dep: libc6 (>= 2.34) [non alpha, ia64, sh4]

  bibliothèque C GNU : bibliothèques partagées
  un paquet virtuel est également fourni par libc6-udeb

  dep: libc6 (>= 2.40) [sh4]

• dep: libc6.1 (>= 2.34) [alpha]

  bibliothèque C GNU : bibliothèques partagées
  un paquet virtuel est également fourni par libc6.1-udeb

  dep: libc6.1 (>= 2.37) [ia64]

• dep: libmpich12 [armel, armhf, hppa, i386, m68k, sh4]

  Shared libraries for MPICH

  dep: libmpich12 (>= 4.2.0) [x32]

• dep: libopenmpi3t64 (>= 4.1.6) [non amd64, armel, armhf, hppa, i386, m68k, sh4, x32]

  high performance message passing library -- shared library

  dep: libopenmpi3t64 (>= 5.0.5) [amd64]

• dep: libsuperlu-dist8 (= 8.2.1+dfsg1-1+b1) [ia64]

  Highly distributed solution of sparse linear equations

  dep: libsuperlu-dist8 (= 8.2.1+dfsg1-2) [alpha, arm64, mips64el, ppc64, ppc64el, riscv64, s390x, sparc64]

  dep: libsuperlu-dist8 (= 8.2.1+dfsg1-5) [amd64, armel, armhf, hppa, i386, m68k, sh4, x32]

Télécharger libsuperlu-dist-dev

Télécharger pour toutes les architectures proposées

Architecture	Version	Taille du paquet	Espace occupé une fois installé	Fichiers
alpha (portage non officiel)	8.2.1+dfsg1-2	927,3 ko	11 411,0 ko	[liste des fichiers]
amd64	8.2.1+dfsg1-5	881,4 ko	7 004,0 ko	[liste des fichiers]
arm64	8.2.1+dfsg1-2	823,6 ko	9 100,0 ko	[liste des fichiers]
armel	8.2.1+dfsg1-5	827,4 ko	5 992,0 ko	[liste des fichiers]
armhf	8.2.1+dfsg1-5	787,5 ko	5 139,0 ko	[liste des fichiers]
hppa (portage non officiel)	8.2.1+dfsg1-5	896,3 ko	6 271,0 ko	[liste des fichiers]
i386	8.2.1+dfsg1-5	854,4 ko	6 322,0 ko	[liste des fichiers]
ia64 (portage non officiel)	8.2.1+dfsg1-1+b1	1 032,7 ko	10 940,0 ko	[liste des fichiers]
m68k (portage non officiel)	8.2.1+dfsg1-5	751,4 ko	5 533,0 ko	[liste des fichiers]
mips64el	8.2.1+dfsg1-2	859,9 ko	10 233,0 ko	[liste des fichiers]
ppc64 (portage non officiel)	8.2.1+dfsg1-2	925,1 ko	9 850,0 ko	[liste des fichiers]
ppc64el	8.2.1+dfsg1-2	952,3 ko	9 704,0 ko	[liste des fichiers]
riscv64	8.2.1+dfsg1-2	2 825,9 ko	29 371,0 ko	[liste des fichiers]
s390x	8.2.1+dfsg1-2	982,6 ko	7 884,0 ko	[liste des fichiers]
sh4 (portage non officiel)	8.2.1+dfsg1-5	1 006,5 ko	7 343,0 ko	[liste des fichiers]
sparc64 (portage non officiel)	8.2.1+dfsg1-2	856,4 ko	47 777,0 ko	[liste des fichiers]
x32 (portage non officiel)	8.2.1+dfsg1-5	858,7 ko	5 984,0 ko	[liste des fichiers]