Compiling Simvascular on EXPANSE (SDSC HPC)

Provides a system for patient-specific cardiovascular modeling and simulation.
POST REPLY
User avatar
Jimmy Azarnoosh
Posts: 25
Joined: Mon Oct 05, 2020 5:43 pm

Compiling Simvascular on EXPANSE (SDSC HPC)

Post by Jimmy Azarnoosh » Thu May 02, 2024 12:33 pm

Hello,

We recently compiled svSolver on EXPANSE (SDSC HPC) and successfully ran jobs on it. We are trying to compile Simvascular as well from the source code in the following link:
https://github.com/SimVascular/SimVascular

All necessary modules that we used to compile svSolver were loaded for Simvascular and CentOS in quick-build-linux.sh was replaced with CentOS|'ROCKY' in line 107.

Executing the code:

Code: Select all

sh quick-build-linux.sh
It failed to properly compile Simvascular on EXPANSE. Please see the compilation output files in the attachment. Output is split because of its size.
The reason we want to compile Simvascular on EXPANSE is to import sv and vtk to our Python codes. If there is any other way around, we would be happy to learn about that.

Any suggestion would be appreciated.


Thank you so much
Jimmy
Attachments
output2.txt
(606.91 KiB) Downloaded 64 times
output1.txt
(505.9 KiB) Downloaded 58 times
Terminal.txt
(451.65 KiB) Downloaded 59 times

User avatar
David Parker
Posts: 1615
Joined: Tue Aug 23, 2005 2:43 pm

Re: Compiling Simvascular on EXPANSE (SDSC HPC)

Post by David Parker » Thu May 02, 2024 1:28 pm

Hi Jimmy,

Building SV on an HPC cluster will be very challenging because a cluster typically does not have the graphics libraries (e.g. OpenGL, X11) needed to build SV. The compilation errors you are seeing are caused by that.

There is a SV CentOS 7 version on SimTK (https://simtk.org/frs/download_confirm. ... oup_id=188) that I built on a CentOS virtual machine, includes copies of the graphics libraries. I don't know if that will work on CentOS 8 or not.

A better solution would be to run SV from within a Docker container running Ubuntu, just need to download and install SV in the container. You could then use the docker exec command to run SV Python scripts in the Docker container. I've not done this before so I don't know all the details but I think it should work.

Cheers,
Dave

POST REPLY