{"schema_version":1,"metadata":{"branch":"current","release":"7.9","architecture":"amd64","generated_at":"2026-09-09T04:04:41.004595+00:00","source_url":"https://cdn.openbsd.org/pub/OpenBSD/snapshots/packages/amd64/sqlports-7.55.tgz","source_sha256":"fcdbba1b9df747882aa9bab4af20ebec8dac10655069c90f1cd4145c06e1235e","package_count":12031,"source_kind":"sqlports"},"package":{"name":"arpack-3.8.0p4v0","path":"math/arpack","url":"/packages/current/math/arpack/","comment":"F77 subroutines for solving large scale eigenvalue problems","homepage":"https://github.com/opencollab/arpack-ng","maintainer":"The OpenBSD ports mailing-list <ports@openbsd.org>","description":"ARPACK is a collection of Fortran77 subroutines designed to solve large\nscale eigenvalue problems.\nIt is a fork of the Rice University ARPACK, that was created as a joint project\nbetween Debian, Octave and Scilab and is now a community project maintained by\na few volunteers.\n\nThe package is designed to compute a few eigenvalues and corresponding\neigenvectors of a general n by n matrix A. It is most appropriate for\nlarge sparse or structured matrices A where structured means that a\nmatrix-vector product w <- Av requires order n rather than the usual\norder n2 floating point operations. This software is based upon an\nalgorithmic variant of the Arnoldi process called the Implicitly\nRestarted Arnoldi Method (IRAM). When the matrix A is symmetric it\nreduces to a variant of the Lanczos process called the Implicitly\nRestarted Lanczos Method (IRLM). These variants may be viewed as a\nsynthesis of the Arnoldi/Lanczos process with the Implicitly Shifted QR\ntechnique that is suitable for large scale problems. For many standard\nproblems, a matrix factorization is not required. Only the action of the\nmatrix on a vector is needed.\n\nARPACK software is capable of solving large scale symmetric,\nnonsymmetric, and generalized eigenproblems from significant application\nareas. The software is designed to compute a few (k) eigenvalues with\nuser specified features such as those of largest real part or largest\nmagnitude. Storage requirements are on the order of n*k locations. No\nauxiliary storage is required. A set of Schur basis vectors for the\ndesired k-dimensional eigen-space is computed which is numerically\northogonal to working precision. Numerically accurate eigenvectors are\navailable on request.\n\nFlavors:\n\tmpi - Build with OpenMPI support\n","package_architecture":"amd64","stem":"arpack","readme":null,"dependencies":[{"path":"math/lapack","type":"library","package_spec":"","url":"/packages/current/math/lapack/"},{"path":"devel/cmake/core","type":"build","package_spec":"STEM->=4","url":"/packages/current/devel/cmake/core/"},{"path":"devel/ninja","type":"build","package_spec":"","url":"/packages/current/devel/ninja/"},{"path":"lang/gcc/16","type":"build","package_spec":"STEM->=16,<17","url":"/packages/current/lang/gcc/16/"},{"path":"lang/gcc/16,-f95","type":"build","package_spec":"STEM->=16,<17","url":"/packages/current/lang/gcc/16,-f95/"}],"reverse_dependencies":{"count":4,"url":"/packages/data/94/94d9905f267a3bc4e747e4402b78ab3fe1935f5dd771e1ab2c895eb014f63087.json"},"categories":["math"],"flavors":["no_mpi"],"only_for_architectures":[],"not_for_architectures":[]}}