You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
psblas3/README.md

121 lines
4.4 KiB
Markdown

PSBLAS library, version 3.5
===========================
10 years ago
7 years ago
The architecture of the Fortran 2003 sparse BLAS is described in:
>S. Filippone, A. Buttari. Object-Oriented Techniques for Sparse Matrix
>Computations in Fortran 2003, ACM Trans. on Math. Software, vol. 38, No.
4, 2012.
The ideas are explored further with the paper:
>V. Cardellini, S. Filippone and D. Rouson. Design Patterns for
>sparse-matrix computations on hybrid CPU/GPU platforms, Scientific
>Programming, 22(2014), pp.1-19.
7 years ago
Version 1.0 of the library is described in:
>S. Filippone, M. Colajanni. PSBLAS: A library for parallel linear
>algebra computation on sparse matrices, ACM Trans. on Math. Software,
>26(4), Dec. 2000, pp. 527-550.
19 years ago
UTILITIES
---------
The `test/util` directory contains some utilities to convert to/from
7 years ago
Harwell-Boeing and MatrixMarket file formats.
19 years ago
DOCUMENTATION
-------------
See docs/psblas-3.5.pdf; an HTML version of the same document is
7 years ago
available in docs/html. Please consult the sample programs, especially
test/pargen/psb_[sd]_pde[23]d.f90
19 years ago
OTHER SOFTWARE CREDITS
----------------------
7 years ago
We originally included a modified implementation of some of the Sparker
(serial sparse BLAS) material; this has been completely rewritten, way
beyond the intention(s) and responsibilities of the original developers.
The main reference for the serial sparse BLAS is:
>Duff, I., Marrone, M., Radicati, G., and Vittoli, C. Level 3 basic
>linear algebra subprograms for sparse matrices: a user level interface,
>ACM Trans. Math. Softw., 23(3), 379-401, 1997.
19 years ago
INSTALLING
----------
To compile and run our software you will need the following
prerequisites (see also SERIAL below):
1. A working version of MPI
7 years ago
2. A version of the BLAS; if you don't have a specific version for your
platform you may try ATLAS available from
http://math-atlas.sourceforge.net/
3. We have had good results with the METIS library, from
7 years ago
http://www-users.cs.umn.edu/~karypis/metis/metis/main.html.
psblas-dev: README base/comm/psb_dgather.f90 base/comm/psb_dhalo.f90 base/comm/psb_dovrl.f90 base/comm/psb_dscatter.F90 base/comm/psb_ihalo.f90 base/comm/psb_zgather.f90 base/comm/psb_zhalo.f90 base/comm/psb_zovrl.f90 base/comm/psb_zscatter.F90 base/internals/psi_dswapdata.F90 base/internals/psi_dswaptran.F90 base/internals/psi_iswapdata.F90 base/internals/psi_iswaptran.F90 base/internals/psi_zswapdata.F90 base/internals/psi_zswaptran.F90 base/modules/Makefile base/modules/cutil.c base/modules/fakempi.c base/modules/psb_comm_mod.f90 base/modules/psb_const_mod.f90 base/modules/psb_desc_type.f90 base/modules/psb_error_mod.F90 base/modules/psb_inter_desc_type.f90 base/modules/psb_penv_mod.F90 base/modules/psb_psblas_mod.f90 base/modules/psb_realloc_mod.F90 base/modules/psb_serial_mod.f90 base/modules/psb_sort_mod.f90 base/modules/psb_spmat_type.f90 base/modules/psb_spsb_mod.f90 base/modules/psb_tools_mod.f90 base/modules/psi_mod.f90 base/modules/psi_serial_mod.f90 base/psblas/psb_damax.f90 base/psblas/psb_dasum.f90 base/psblas/psb_daxpby.f90 base/psblas/psb_ddot.f90 base/psblas/psb_dnrm2.f90 base/psblas/psb_dnrmi.f90 base/psblas/psb_dspmm.f90 base/psblas/psb_dspsm.f90 base/psblas/psb_zamax.f90 base/psblas/psb_zasum.f90 base/psblas/psb_zaxpby.f90 base/psblas/psb_zdot.f90 base/psblas/psb_znrm2.f90 base/psblas/psb_znrmi.f90 base/psblas/psb_zspmm.f90 base/psblas/psb_zspsm.f90 base/serial/aux/dasr.f90 base/serial/aux/dasrx.f90 base/serial/aux/dmsort_dw.f90 base/serial/aux/dmsort_up.f90 base/serial/aux/dmsr.f90 base/serial/aux/dmsrx.f90 base/serial/aux/dsr.f90 base/serial/aux/dsrx.f90 base/serial/aux/zacmp_mod.f90 base/serial/aux/zalcmp_mod.f90 base/serial/aux/zalsr.f90 base/serial/aux/zalsrx.f90 base/serial/aux/zamsort_dw.f90 base/serial/aux/zamsort_up.f90 base/serial/aux/zamsr.f90 base/serial/aux/zamsrx.f90 base/serial/aux/zasr.f90 base/serial/aux/zasrx.f90 base/serial/aux/zlcmp_mod.f90 base/serial/aux/zlsr.f90 base/serial/aux/zlsrx.f90 base/serial/coo/dcoomm.f base/serial/coo/dcoomv.f base/serial/coo/dcoonrmi.f base/serial/coo/dcoorws.f base/serial/coo/dcoosm.f base/serial/coo/dcoosv.f base/serial/coo/zcoomm.f base/serial/coo/zcoomv.f base/serial/coo/zcoonrmi.f base/serial/coo/zcoorws.f base/serial/coo/zcoosm.f base/serial/coo/zcoosv.f base/serial/csr/dcrnrmi.f base/serial/csr/dcsrck.f base/serial/csr/dcsrmm.f base/serial/csr/dcsrmv.f base/serial/csr/dcsrmv2.f base/serial/csr/dcsrmv3.f base/serial/csr/dcsrmv4.f base/serial/csr/dcsrrws.f base/serial/csr/dcsrsm.f base/serial/csr/dcsrsv.f base/serial/csr/zcrnrmi.f base/serial/csr/zcsrck.f base/serial/csr/zcsrmm.f base/serial/csr/zcsrrws.f base/serial/csr/zcsrsm.f base/serial/csr/zsrmv.f base/serial/csr/zsrsv.f base/serial/dp/dcoco.f base/serial/dp/dcocr.f base/serial/dp/dcrco.f base/serial/dp/dcrcr.f base/serial/dp/dcrjd.f base/serial/dp/dcsrp1.f base/serial/dp/dcsrrp.f base/serial/dp/dgind_tri.f base/serial/dp/dgindex.f base/serial/dp/djadrp.f base/serial/dp/djadrp1.f base/serial/dp/djdco.f base/serial/dp/djdcox.f base/serial/dp/dvtfg.f base/serial/dp/reordvn.f base/serial/dp/zcoco.f base/serial/dp/zcocr.f base/serial/dp/zcrco.f base/serial/dp/zcrcr.f base/serial/dp/zcrjd.f base/serial/dp/zgind_tri.f base/serial/dp/zgindex.f base/serial/f77/daxpby.f base/serial/f77/dcsmm.f base/serial/f77/dcsnmi.f base/serial/f77/dcsrp.f base/serial/f77/dcsrws.f base/serial/f77/dcssm.f base/serial/f77/dgelp.f base/serial/f77/dlpupd.f base/serial/f77/dswmm.f base/serial/f77/dswsm.f base/serial/f77/smmp.f base/serial/f77/zaxpby.f base/serial/f77/zcsmm.f base/serial/f77/zcsnmi.f base/serial/f77/zcsrws.f base/serial/f77/zcssm.f base/serial/f77/zgelp.f base/serial/f77/zlpupd.f base/serial/f77/zswmm.f base/serial/f77/zswsm.f base/serial/jad/djadmm.f base/serial/jad/djadmv.f base/serial/jad/djadmv2.f base/serial/jad/djadmv3.f base/serial/jad/djadmv4.f base/serial/jad/djadnr.f base/serial/jad/djadrws.f base/serial/jad/djadsm.f base/serial/jad/djadsv.f base/serial/jad/djdnrmi.f base/serial/jad/djdrws.f base/serial/psb_dcoins.f90 base/serial/psb_dcsmm.f90 base/serial/psb_dcsmv.f90 base/serial/psb_dcsnmi.f90 base/serial/psb_dcsrp.f90 base/serial/psb_dcsrws.f90 base/serial/psb_dcssm.f90 base/serial/psb_dcssv.f90 base/serial/psb_dgelp.f90 base/serial/psb_dneigh.f90 base/serial/psb_dnumbmm.f90 base/serial/psb_dspcnv.f90 base/serial/psb_dspgetrow.f90 base/serial/psb_dspgtdiag.f90 base/serial/psb_dspscal.f90 base/serial/psb_dsymbmm.f90 base/serial/psb_getrow_mod.f90 base/serial/psb_regen_mod.f90 base/serial/psb_update_mod.f90 base/serial/psb_zcoins.f90 base/serial/psb_zcsmm.f90 base/serial/psb_zcsmv.f90 base/serial/psb_zcsnmi.f90 base/serial/psb_zcsrp.f90 base/serial/psb_zcsrws.f90 base/serial/psb_zcssm.f90 base/serial/psb_zcssv.f90 base/serial/psb_zgelp.f90 base/serial/psb_zneigh.f90 base/serial/psb_znumbmm.f90 base/serial/psb_zspcnv.f90 base/serial/psb_zspgetrow.f90 base/serial/psb_zspgtdiag.f90 base/serial/psb_zspscal.f90 base/serial/psb_zsymbmm.f90 base/tools/psb_dallc.f90 base/tools/psb_dasb.f90 base/tools/psb_dfree.f90 base/tools/psb_dins.f90 base/tools/psb_dsphalo.F90 base/tools/psb_dspins.f90 base/tools/psb_zallc.f90 base/tools/psb_zasb.f90 base/tools/psb_zfree.f90 base/tools/psb_zins.f90 base/tools/psb_zsphalo.F90 base/tools/psb_zspins.f90 docs/pdf/datastruct.tex docs/pdf/penv.tex docs/userguide.pdf krylov/psb_dbicg.f90 krylov/psb_dcg.f90 krylov/psb_dcgs.f90 krylov/psb_dcgstab.F90 krylov/psb_dcgstabl.f90 krylov/psb_drgmres.f90 krylov/psb_krylov_mod.f90 krylov/psb_zbicg.f90 krylov/psb_zcg.f90 krylov/psb_zcgs.f90 krylov/psb_zcgstab.f90 krylov/psb_zcgstabl.f90 krylov/psb_zrgmres.f90 prec/psb_dbjac_aply.f90 prec/psb_dbjac_bld.f90 prec/psb_dgprec_aply.f90 prec/psb_dilu_fct.f90 prec/psb_dprc_aply.f90 prec/psb_dprecset.f90 prec/psb_prec_mod.f90 prec/psb_prec_type.f90 prec/psb_zbjac_aply.f90 prec/psb_zbjac_bld.f90 prec/psb_zgprec_aply.f90 prec/psb_zilu_fct.f90 prec/psb_zprc_aply.f90 prec/psb_zprecset.f90 test/fileread/df_sample.f90 test/fileread/getp.f90 test/fileread/zf_sample.f90 test/pargen/ppde.f90 test/util/Makefile util/psb_hbio_mod.f90 util/psb_mat_dist_mod.f90 util/psb_metispart_mod.F90 util/psb_mmio_mod.f90 util/psb_read_mat_mod.f90 Merged kind type names changes from r:2905:2906 in branches/psblas-2.2-maint.
17 years ago
This is optional; it is used in the util and test/fileread
directories but only if you specify `--with-metis`.
4. If you have the AMD package of Davis, Duff and Amestoy, you can
specify `--with-amd` (see `./configure --help` for more details).
7 years ago
The configure script will generate a Make.inc file suitable for building
the library. The script is capable of recognizing the needed libraries
with their default names; if they are in unusual places consider adding
the paths with `--with-libs`, or explicitly specifying the names in
`--with-blas`, etc. Please note that a common way for the configure script
7 years ago
to fail is to specify inconsistent MPI vs. plain compilers, either
directly or indirectly via environment variables; e.g. specifying the
Intel compiler with `FC=ifort` while at the same time having an
`MPIFC=mpif90` which points to GNU Fortran. The best way to avoid this
7 years ago
situation is (in our opinion) to use the environment modules package
(see http://modules.sourceforge.net/), and load the relevant
variables with (e.g.)
```
7 years ago
module load gnu46 openmpi
```
7 years ago
This will delegate to the modules setup to make sure that the version of
openmpi in use is the one compiled with the gnu46 compilers. After the
configure script has completed you can always tweak the Make.inc file
yourself.
psblas-dev: README base/comm/psb_dgather.f90 base/comm/psb_dhalo.f90 base/comm/psb_dovrl.f90 base/comm/psb_dscatter.F90 base/comm/psb_ihalo.f90 base/comm/psb_zgather.f90 base/comm/psb_zhalo.f90 base/comm/psb_zovrl.f90 base/comm/psb_zscatter.F90 base/internals/psi_dswapdata.F90 base/internals/psi_dswaptran.F90 base/internals/psi_iswapdata.F90 base/internals/psi_iswaptran.F90 base/internals/psi_zswapdata.F90 base/internals/psi_zswaptran.F90 base/modules/Makefile base/modules/cutil.c base/modules/fakempi.c base/modules/psb_comm_mod.f90 base/modules/psb_const_mod.f90 base/modules/psb_desc_type.f90 base/modules/psb_error_mod.F90 base/modules/psb_inter_desc_type.f90 base/modules/psb_penv_mod.F90 base/modules/psb_psblas_mod.f90 base/modules/psb_realloc_mod.F90 base/modules/psb_serial_mod.f90 base/modules/psb_sort_mod.f90 base/modules/psb_spmat_type.f90 base/modules/psb_spsb_mod.f90 base/modules/psb_tools_mod.f90 base/modules/psi_mod.f90 base/modules/psi_serial_mod.f90 base/psblas/psb_damax.f90 base/psblas/psb_dasum.f90 base/psblas/psb_daxpby.f90 base/psblas/psb_ddot.f90 base/psblas/psb_dnrm2.f90 base/psblas/psb_dnrmi.f90 base/psblas/psb_dspmm.f90 base/psblas/psb_dspsm.f90 base/psblas/psb_zamax.f90 base/psblas/psb_zasum.f90 base/psblas/psb_zaxpby.f90 base/psblas/psb_zdot.f90 base/psblas/psb_znrm2.f90 base/psblas/psb_znrmi.f90 base/psblas/psb_zspmm.f90 base/psblas/psb_zspsm.f90 base/serial/aux/dasr.f90 base/serial/aux/dasrx.f90 base/serial/aux/dmsort_dw.f90 base/serial/aux/dmsort_up.f90 base/serial/aux/dmsr.f90 base/serial/aux/dmsrx.f90 base/serial/aux/dsr.f90 base/serial/aux/dsrx.f90 base/serial/aux/zacmp_mod.f90 base/serial/aux/zalcmp_mod.f90 base/serial/aux/zalsr.f90 base/serial/aux/zalsrx.f90 base/serial/aux/zamsort_dw.f90 base/serial/aux/zamsort_up.f90 base/serial/aux/zamsr.f90 base/serial/aux/zamsrx.f90 base/serial/aux/zasr.f90 base/serial/aux/zasrx.f90 base/serial/aux/zlcmp_mod.f90 base/serial/aux/zlsr.f90 base/serial/aux/zlsrx.f90 base/serial/coo/dcoomm.f base/serial/coo/dcoomv.f base/serial/coo/dcoonrmi.f base/serial/coo/dcoorws.f base/serial/coo/dcoosm.f base/serial/coo/dcoosv.f base/serial/coo/zcoomm.f base/serial/coo/zcoomv.f base/serial/coo/zcoonrmi.f base/serial/coo/zcoorws.f base/serial/coo/zcoosm.f base/serial/coo/zcoosv.f base/serial/csr/dcrnrmi.f base/serial/csr/dcsrck.f base/serial/csr/dcsrmm.f base/serial/csr/dcsrmv.f base/serial/csr/dcsrmv2.f base/serial/csr/dcsrmv3.f base/serial/csr/dcsrmv4.f base/serial/csr/dcsrrws.f base/serial/csr/dcsrsm.f base/serial/csr/dcsrsv.f base/serial/csr/zcrnrmi.f base/serial/csr/zcsrck.f base/serial/csr/zcsrmm.f base/serial/csr/zcsrrws.f base/serial/csr/zcsrsm.f base/serial/csr/zsrmv.f base/serial/csr/zsrsv.f base/serial/dp/dcoco.f base/serial/dp/dcocr.f base/serial/dp/dcrco.f base/serial/dp/dcrcr.f base/serial/dp/dcrjd.f base/serial/dp/dcsrp1.f base/serial/dp/dcsrrp.f base/serial/dp/dgind_tri.f base/serial/dp/dgindex.f base/serial/dp/djadrp.f base/serial/dp/djadrp1.f base/serial/dp/djdco.f base/serial/dp/djdcox.f base/serial/dp/dvtfg.f base/serial/dp/reordvn.f base/serial/dp/zcoco.f base/serial/dp/zcocr.f base/serial/dp/zcrco.f base/serial/dp/zcrcr.f base/serial/dp/zcrjd.f base/serial/dp/zgind_tri.f base/serial/dp/zgindex.f base/serial/f77/daxpby.f base/serial/f77/dcsmm.f base/serial/f77/dcsnmi.f base/serial/f77/dcsrp.f base/serial/f77/dcsrws.f base/serial/f77/dcssm.f base/serial/f77/dgelp.f base/serial/f77/dlpupd.f base/serial/f77/dswmm.f base/serial/f77/dswsm.f base/serial/f77/smmp.f base/serial/f77/zaxpby.f base/serial/f77/zcsmm.f base/serial/f77/zcsnmi.f base/serial/f77/zcsrws.f base/serial/f77/zcssm.f base/serial/f77/zgelp.f base/serial/f77/zlpupd.f base/serial/f77/zswmm.f base/serial/f77/zswsm.f base/serial/jad/djadmm.f base/serial/jad/djadmv.f base/serial/jad/djadmv2.f base/serial/jad/djadmv3.f base/serial/jad/djadmv4.f base/serial/jad/djadnr.f base/serial/jad/djadrws.f base/serial/jad/djadsm.f base/serial/jad/djadsv.f base/serial/jad/djdnrmi.f base/serial/jad/djdrws.f base/serial/psb_dcoins.f90 base/serial/psb_dcsmm.f90 base/serial/psb_dcsmv.f90 base/serial/psb_dcsnmi.f90 base/serial/psb_dcsrp.f90 base/serial/psb_dcsrws.f90 base/serial/psb_dcssm.f90 base/serial/psb_dcssv.f90 base/serial/psb_dgelp.f90 base/serial/psb_dneigh.f90 base/serial/psb_dnumbmm.f90 base/serial/psb_dspcnv.f90 base/serial/psb_dspgetrow.f90 base/serial/psb_dspgtdiag.f90 base/serial/psb_dspscal.f90 base/serial/psb_dsymbmm.f90 base/serial/psb_getrow_mod.f90 base/serial/psb_regen_mod.f90 base/serial/psb_update_mod.f90 base/serial/psb_zcoins.f90 base/serial/psb_zcsmm.f90 base/serial/psb_zcsmv.f90 base/serial/psb_zcsnmi.f90 base/serial/psb_zcsrp.f90 base/serial/psb_zcsrws.f90 base/serial/psb_zcssm.f90 base/serial/psb_zcssv.f90 base/serial/psb_zgelp.f90 base/serial/psb_zneigh.f90 base/serial/psb_znumbmm.f90 base/serial/psb_zspcnv.f90 base/serial/psb_zspgetrow.f90 base/serial/psb_zspgtdiag.f90 base/serial/psb_zspscal.f90 base/serial/psb_zsymbmm.f90 base/tools/psb_dallc.f90 base/tools/psb_dasb.f90 base/tools/psb_dfree.f90 base/tools/psb_dins.f90 base/tools/psb_dsphalo.F90 base/tools/psb_dspins.f90 base/tools/psb_zallc.f90 base/tools/psb_zasb.f90 base/tools/psb_zfree.f90 base/tools/psb_zins.f90 base/tools/psb_zsphalo.F90 base/tools/psb_zspins.f90 docs/pdf/datastruct.tex docs/pdf/penv.tex docs/userguide.pdf krylov/psb_dbicg.f90 krylov/psb_dcg.f90 krylov/psb_dcgs.f90 krylov/psb_dcgstab.F90 krylov/psb_dcgstabl.f90 krylov/psb_drgmres.f90 krylov/psb_krylov_mod.f90 krylov/psb_zbicg.f90 krylov/psb_zcg.f90 krylov/psb_zcgs.f90 krylov/psb_zcgstab.f90 krylov/psb_zcgstabl.f90 krylov/psb_zrgmres.f90 prec/psb_dbjac_aply.f90 prec/psb_dbjac_bld.f90 prec/psb_dgprec_aply.f90 prec/psb_dilu_fct.f90 prec/psb_dprc_aply.f90 prec/psb_dprecset.f90 prec/psb_prec_mod.f90 prec/psb_prec_type.f90 prec/psb_zbjac_aply.f90 prec/psb_zbjac_bld.f90 prec/psb_zgprec_aply.f90 prec/psb_zilu_fct.f90 prec/psb_zprc_aply.f90 prec/psb_zprecset.f90 test/fileread/df_sample.f90 test/fileread/getp.f90 test/fileread/zf_sample.f90 test/pargen/ppde.f90 test/util/Makefile util/psb_hbio_mod.f90 util/psb_mat_dist_mod.f90 util/psb_metispart_mod.F90 util/psb_mmio_mod.f90 util/psb_read_mat_mod.f90 Merged kind type names changes from r:2905:2906 in branches/psblas-2.2-maint.
17 years ago
After you have Make.inc fixed, run
```
7 years ago
make
```
7 years ago
to compile the library; go to the test directory and its subdirectories
to get test programs done. If you specify `--prefix=/path` you can do make
install and the libraries will be installed under `/path/lib`, while the
module files will be installed under `/path/include`.
SERIAL
------
Configuring with `--enable-serial` will provide a fake MPI stub library
7 years ago
that enables running in pure serial mode; no MPI installation is needed
in this case (but note that the fake MPI stubs are only guaranteed to
cover what we use internally, it's not a complete replacement).
LONG INTEGERS
-------------
We have an experimental flag `--enable-long-integers` that will enable
7 years ago
having 8-byte integer data, allowing an index space larger than 2G; some
small cases have been tested but we do not offer full guarantee (yet).
19 years ago
TODO
----
Fix all reamining bugs. Bugs? We dont' have any ! ;-)
19 years ago
The PSBLAS team.
RELATED SOFTWARE
----------------
If you are looking for more sophisticated preconditioners, you may be
interested in the package MLD2P4 from
<http://github.com/sfilippone/mld2p4-2>
19 years ago
Contact: <https://github.com/sfilippone/psblas3>