Skip to main content

Table 2 Total runtime depending on the number of MPI processes, and time spent in the FFT calls for the standard CG algorithm. The second column lists the number of slices per MPI process

From: Runtime optimization of a memory efficient CG solver for FFT-based homogenization: implementation details and scaling results for linear elasticity

Processes Slices Total time \((\mathrm{{s}})\) FFT (s) FFT (s)
8 64 2239.87 1081.80 48
16 32 1150.57 558.29 48
32 16 636.72 326.91 51
64 8 509.52 281.86 49
128 4 289.07 153.50 55
256 2 170.80 90.35 53