Table 2
Numerical simulation computation times.
Case | τ(LareJ) | τ(HYDRAD) | τ(Lare1D(8000)) | τ(HYDRAD) / | τ(Lare1D(8000)) / |
(min) | (min) | (min) | τ(LareJ) | τ(LareJ) | |
|
|||||
1 | 17 | 316 | 7426 | 18.6 | 436.8 |
2 | 19 | 340 | 7766 | 17.9 | 408.7 |
3 | 51 | 1943 | 13 886 | 38.1 | 272.3 |
4 | 22 | 370 | 6341 | 16.8 | 288.2 |
5 | 82 | 2617 | 8594* | 31.9 | 106.0 |
6 | 154 | 5177 | 12 732* | 33.6 | 82.7 |
7 | 26 | 1559 | 18 893 | 60.0 | 726.7 |
8 | 28 | 1566 | 18 059 | 56.0 | 645.0 |
9 | 35 | 1605 | 16 833 | 45.9 | 480.9 |
10 | 26 | 1805 | 11 138 | 69.4 | 428.4 |
11 | 32 | 1914 | 11 997 | 59.8 | 374.9 |
12 | 86 | 2269 | 12 973* | 26.4 | 150.8 |
Notes. The columns show the computation times (run on a single processor) using the Lare1D code with 500 grid points (coarse resolution) employed with the jump condition (LareJ), the HYDRAD code (in single fluid mode) with the largest grid cell of width 400 km and 12 levels of refinement employed, the Lare1D code using 8000 grid points along the length of the loop (Lare1D(8000)), and the computational time ratios between these methods. The short loop simulations (Cases 1−6) are run to a final time of 4000 s and the long loop simulations (Cases 7−12) are run to a final time of 12 000 s. The asterisks indicate cases where the Lare1D code using 8000 grid points was unable to resolve the density to within 75% of the HYDRAD solution.
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.