https://ww8central.ww.uni-erlangen.de/ 2026-04-10T10:28:57+00:00 https://ww8central.ww.uni-erlangen.de/ https://ww8central.ww.uni-erlangen.de/_media/wiki/logo.png text/html 2014-08-12T14:37:39+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/software/scientific_libraries/blas_lapack?rev=1407854259&do=diff BLAS/LAPACK BLAS stands for Basic Linear Algebra Subprograms. This library contains linear algebraic operations that include matrixes and vectors. Its functions are separated into three parts: * Level 1 routines to represent vectors and vector/vector operations. text/html 2016-10-05T19:00:03+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/software/scientific_libraries/fftw?rev=1475694003&do=diff FFTW in parallel --enable-threads --with-openmp <http://www.fftw.org/fftw2_doc/fftw_4.html> fftw_threads_test 2 -s 128x128 Test : best number of threads MPI mpi enabled version of fftw required (no Fedora system lib available) --enable-mpi The order of the linked libraries is impotent as always: text/html 2019-11-25T13:28:41+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/software/scientific_libraries/intelmkl?rev=1574688521&do=diff MKL To get the compiler and linker options for using the MKL you can use the Intel® Math Kernel Library Link Line Advisor sim04/sim05 As on all non-intel chips, MKL will not automagically select the correct code path for fastest exectuion. The same holds true for multi-code-path (fat) binaries generated by the intel compilers (ifort/icc). text/html 2015-10-12T10:56:19+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/software/scientific_libraries/openblas?rev=1444647379&do=diff OpenBLAS/GotoBLAS Website <http://www.openblas.net/> OpenBLAS 0.2.14 Fedora 22 : -system included Environment variables for development module load 0.2.14-system OpenBLAS is an optimized BLAS library based on GotoBLAS2 1.13 BSD version. The project is supported by the Lab of Parallel Software and Computational Science, ISCAS. text/html 2014-08-12T14:37:39+00:00 Anonymous (anonymous@undisclosed.example.com) https://ww8central.ww.uni-erlangen.de/software/scientific_libraries/scalapack?rev=1407854259&do=diff SCALAPACK SCAlable Linear Algebra PACKage This library is the scalable version of LAPACK. Both libraries use block partitioning to reduce data exchanges between the different memory levels to a minimum. SCALAPACK is above all used for eigenvalue problems and factorizations (LU, Cholesky and QR).