Closed krober10nd closed 3 years ago
Updating the parallel performance figure in accord with #117
Added a larger benchmark and replaced the picture of the mesh with a normalized mesh generation rate.
Updated the text in the paper accordingly.
Primary slow down is a combination of load imbalance and axpy and array manipulation which happens in Python.
Updating the parallel performance figure in accord with #117
Added a larger benchmark and replaced the picture of the mesh with a normalized mesh generation rate.
Updated the text in the paper accordingly.
Primary slow down is a combination of load imbalance and axpy and array manipulation which happens in Python.