====== ELPA ======

Descargar un .tar de la web de ELPA.
https://elpa.mpcdf.mpg.de/software/tarball-archive/Releases

Lo descomprimimos en /data/shared/apps/elpa/version-number.

Con este enviroment:

<code>Currently Loaded Modules:
  1) tbb/2021.10.0          3) oclfpga/2023.2.1    5) intel/2023.2.1   7) ucx/1.15.0         9) openmpi4/4.1.6
  2) compiler-rt/2023.2.1   4) compiler/2023.2.1   6) hwloc/2.7.2      8) libfabric/1.19.0  10) mkl/2023.2.0
</code>

Hacemos el siguiente configure dentro de un nodo de cómputo:

<code>./configure --prefix=/data/shared/apps/elpa/2025-01-002/build \
--enable-openmp --enable-avx2 --enable-avx512 \
CC=mpicc FC=mpif90 CXX=mpicxx \
FCFLAGS="-O3 -xCORE-AVX512" CFLAGS="-O3 -xCORE-AVX512" \
CXXFLAGS="-O3 -xCORE-AVX512 -I/opt/ohpc/pub/mpi/openmpi4-intel/4.1.6/include" \
SCALAPACK_LDFLAGS="-L$MKL_HOME/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -lmkl_blacs_intelmpi_lp64 -lpthread -lm -Wl,-rpath,$MKL_HOME/lib/intel64" \
SCALAPACK_FCFLAGS="-L$MKL_HOME/lib/intel64 -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -lmkl_blacs_intelmpi_lp64 -lpthread -lm -I$MKL_HOME/include/intel64/lp64" \
--enable-runtime-threading-support-checks \
--enable-allow-thread-limiting \
--without-threading-support-check-during-build</code>

Luego:

<code>make -j64
make install</code>