On UBUNTU 18.04 LTS and using Anaconda's Python distribution and package management system:

1) Make sure you have OpenMP support (f.e. "sudo apt-get install libopenmp-dev")

2) Install MPI and trilinos library which comes with Zoltan, following PyZoltan doc:
   (https://pyzoltan.readthedocs.io/en/stable/start.html#setting-up-on-ubuntu)
	
	sudo apt-get install openmpi-bin libopenmpi-dev libtrilinos-zoltan-dev

3) Export Zoltan variables before installing PyZoltan and PySPH (make sure the directories match in your system):

	export ZOLTAN_INCLUDE=/usr/include/trilinos
	export ZOLTAN_LIBRARY=/usr/lib/x86_64-linux-gnu
	export USE_TRILINOS=1

4) Create your venv (if you'd like) and follow the instructions to install the core dependencies on Anaconda of PySPH:
   (https://pysph.readthedocs.io/en/latest/installation.html#using-anaconda)

	conda install -c conda-forge cython mako matplotlib jupyter pyside pytest mock

Note: I also added other packages to be safe: numpy-stl, make, nose, pytools

5) Install mpi4py:

	conda install -c conda-forge mpi4py

6) Install CyArray:

	pip install cyarray

7) Install PyZoltan:
	
	pip install pyzoltan

8) Finally, install PySPH:

	pip install --no-cache-dir pysph

9) You should be able to run "pysph test -v" succesfully and have access to CPU parallel capabilities of PySPH.