HemeLB_GPU on JUWELS/V100 performance assessment (POP2_AR_065)

This audit examined strong scaling of a GPU-enabled development version of the HemeLB application on up to 32 of the JUWELS quad-V100 accelerated compute nodes using a (patched) "arteries' geometry dataset

HemeLB_GPU is an open-source lattice-Boltzmann code for simulation of large-scale fluid flow in complex sparse geometries such as those found in vascular networks.  It is written in C++ using MPI+CUDA by UCL and developed within the EU H2020 HPC Centre of Excellence for Computational Biomedicine (CompBioMed).