Open Access

Table B.1

Numerical simulation computation times (run on a single processor) for three different methods to treat thermal conduction.

Case N z τ sts τ cyc τ exp τcyc/ τexp/
(min) (min) (min) τ sts τ sts

1 500 2.45 1.98 2.25 0.81 0.92
1000 6.73 6.47 15.72 0.96 2.34
2000 12.23 29.07 128 2.38 10.5
4000 42.6 199 592 4.67 13.9
8000 205 1537 4699 7.50 22.9

2 500 6.32 8.12 25.7 1.28 4.07
1000 18.5 45.02 122 2.43 6.59
2000 48.8 308 970 6.31 19.9
4000 135 2385 7772 17.7 57.6
8000 607 18 778 47 123* 30.9 77.6

3 500 12.15 33.13 168 2.73 13.8
1000 49.67 257 790 5.17 15.9
2000 138 2023 6238 14.7 45.2
4000 579 15 958 48 405* 27.6 83.6
8000 2440 108 898* 238 620* 44.6 97.8

Notes. The columns show the number of grid points (uniform grid used), the computation times by treating thermal conduction using super time stepping methods (sts), explicit time step sub-cycling (cyc) and explicit time step evolution (exp), and the computation time ratios between these methods. The simulations (Cases 13 of Table 1) are run to a final time of 60 s, which coincides with the end of the heating period. The asterisks indicate runs where the computation time to the final time has been estimated based on results over a shorter period.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.