Closed maho3 closed 4 months ago
To build pinnochio on infinity, first load openmpi
module load openmpi/5.0.3-gnu
After cloning pinnochio from https://github.com/pigimonaco/Pinocchio, edit the Makefile to be compatible with infinity:
ifeq ($(SYSTYPE),"infinity")
CC = mpicc
CDEBUG = -ggdb3 -Wall
COPTIMIZED = -O3 -Wno-unused-result
FFTW_LIBR = -L/softs/fftw3/3.3.10-gnu-mpi/lib -lfftw3_mpi -lfftw3
FFTW_INCL = -I/softs/fftw3/3.3.10-gnu-mpi/include
MPI_LIBR = -L/softs/openmpi/5.0.3-gnu/include
MPI_INCL =
GSL_LIBR = -lgsl -lgslcblas -lm
GSL_INCL = -I/usr/include
endif
Also change the SYSTYPE
argument to be "infinity"
and ensure the "-DWHITENOISE" option is enabled. Building with make clean; make
should then work. Before running the executable pinnochio.x
, you must also run
export LD_LIBRARY_PATH=/softs/fftw3/3.3.10-gnu-mpi/lib:$LD_LIBRARY_PATH
to enable pinnochio to access the fftw3_mpi library.
Pinocchio implemented.
This task is to investigate the following mock generation codes for their possible utility as a baseline for Go Big:
These are used in Euclid and DESI (?), respectively. We should figure out to what halo resolution they resolve, and how fast they are relative to PM simulations.