Gridap (Grid-based approximation of partial differential equations in Julia) plugin to use PETSC (Portable, Extensible Toolkit for Scientific Computation).

using MPI
using Gridap
using GridapPETSC

MPI.Init()
GridapPETSc.Init()

A = sparse([1,2,3,4,5],[1,2,3,4,5],[1.0,2.0,3.0,4.0,5.0])
b = ones(A.n)
x = similar(b)
ps = PETScSolver()
ss = symbolic_setup(ps, A)
ns = numerical_setup(ss, A)
solve!(x, ns, b)

GridapPETSc.Finalize()
MPI.Finalize()

using Gridap
using GridapPETSc

MPI.Init()
GridapPETSc.Init()

# Define the FE problem
# -Δu = x*y in (0,1)^3, u = 0 on the boundary.

model = CartesianDiscreteModel((0,1,0,1,0,1), (10,10,10))

V = TestFESpace(reffe=:Lagrangian, order=1, valuetype=Float64,
  conformity=:H1, model=model, dirichlet_tags="boundary")

U = TrialFESpace(V)

trian = get_triangulation(model)
quad = CellQuadrature(trian,2)

t_Ω = AffineFETerm(
  (u,v) -> inner(∇(v),∇(u)),
  (v) -> inner(v, (x) -> x[1]*x[2] ),
  trian, quad)

op = AffineFEOperator(SparseMatrixCSR{0,PetscReal,PetscInt},U,V,t_Ω)

ls = PETScSolver()
solver = LinearFESolver(ls)

uh = solve(solver,op)

GridapPETSc.Finalize()
MPI.Finalize()

GridPETSc itself is installed when you add and use it into another project.

Please, ensure that your system fulfill the requirements.

To include into your project form Julia REPL, use the following commands:

pkg> add GridapPETSc
julia> using GridapPETSc

If, for any reason, you need to manually build the project, write down the following commands in Julia REPL:

pkg> add GridapPETSc
pkg> build GridPETSc
julia> using GridapPETSc

GridapPETSc julia package requires PETSC library (Portable, Extensible Toolkit for Scientific Computation) and OPENMPI to work correctly. PETSc library can be manually installed in any path on your local machine. In order to succesfull describe your custom installation to be located by GridapPETSc, you must export PETSC_DIR and PETSC_ARCH environment variables. If this environment variables are not available, GridapPETSc will try to find the library in the usual linux user library directory (/usr/lib).

PETSC_DIR and PETSC_ARCH are a couple of variables that control the configuration and build process of PETSc:

• PETSC_DIR: this variable should point to the location of the PETSc installation that is used. Multiple PETSc versions can coexist on the same file-system. By changing PETSC_DIR value, one can switch between these installed versions of PETSc.
• PETSC_ARCH: this variable gives a name to a configuration/build. Configure uses this value to stores the generated config makefiles in ${PETSC_DIR}/${PETSC_ARCH}. Make uses this value to determine this location of these makefiles which intern help in locating the correct include and library files.

Thus one can install multiple variants of PETSc libraries - by providing different PETSC_ARCH values to each configure build. Then one can switch between using these variants of libraries from make by switching the PETSC_ARCH value used.

If configure doesn't find a PETSC_ARCH value (either in env variable or command line option), it automatically generates a default value and uses it. Also - if make doesn't find a PETSC_ARCH env variable - it defaults to the value used by last successful invocation of previous configure. PETSC_ARCH value can be an empty string too.

PETSc can be obtained from the default repositories of your Debian-based OS by means of apt tool.

Basic PETSc installation in order to use it from GridapPETSc julia package is as follows:

$ sudo apt-get update
$ sudo apt-get openmpi petsc-dev

In order to take advantage of GridapPETSc julia package during continuous integration, you must ensure that the requirements are fullfilled in the CI environment.

If your CI process is based on Travis-CI you can add the following block at the beginning of your .travis.yml file:

addons:
  apt:
    update: true
    packages:
    - openmpi-bin
    - petsc-dev

GridapPETSc default sparse matrix format is 0-based compressed sparse row. This types of sparse matrix can be described by SparseMatrixCSR{0,PetscReal,PetscInt} and SymSparseMatrixCSR{0,PetscReal,PetscInt}.These types of matrix are implemented in the SparseMatricesCSR) julia package.