Skip to content

Commit

Permalink
ParU: readme
Browse files Browse the repository at this point in the history
  • Loading branch information
DrTimothyAldenDavis committed Dec 26, 2023
1 parent 13da87c commit f9feb51
Showing 1 changed file with 18 additions and 9 deletions.
27 changes: 18 additions & 9 deletions ParU/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,19 +15,28 @@ Stay tuned.

## Introduction

ParU: is a set of routings for solving sparse linear system via parallel
multifrontal LU factorization algorithms. Requires OpenMP 4.0+, BLAS, CHOLMOD,
ParU is a set of routines for solving sparse linear system via parallel
multifrontal LU factorization algorithms. Requires OpenMP 4.5+, BLAS, CHOLMOD,
UMFPACK, AMD, COLAMD, CAMD, CCOLAMD, and METIS (in particular, the
CHOLMOD/SuiteSparse_metis variant; see the CHOLMOD documentation for details).
`CHOLMOD/SuiteSparse_metis` variant; see the CHOLMOD documentation for
details).

## How to install

You should first install all dependencies for ParU which is UMFPACK and all its
dependencies (AMD, CHOLMOD, ...). By default ParU also needs metis. The
configuration of ParU is mostly done via SuiteSparse config file (BLAS library,
OpenMP settings and ...) which is in SuiteSparse/SuiteSparse_config.
All SuiteSparse dependencies should be in the same directory as in ParU.
After that simply call make.
See the SuiteSparse/README.md for instructions on building all of SuiteSparse
via the SuiteSparse/CMakeLists.txt file. You may also build each individual
package that ParU depends on (`SuiteSparse_config`, AMD, COLAMD, CCAMD,
CCOLAMD, CHOLMOD, and UMFPACK). Then simply do:

```
cd ParU/build
cmake ..
cmake --build . --config Release
sudo cmake --install .
```

Alternatively, on Linux, Mac, or MINGW, simply type `make` in the ParU
folder, then `sudo make install`.

## How to use

Expand Down

0 comments on commit f9feb51

Please sign in to comment.