Open Access

Table 1.

Computing time of the OP grad parallel and sequential runs (‘OP’ and ‘OP seq.’, respectively), and of the sequential OP, SVP2020, and SVP2004 runs scaled to the duration of the parallel OP run (‘OP/OP seq.’, ‘SVP20xx/OP’; xx = 20 or 04 for SVP2020 and SVP2004, respectively) for each kind of model.

Type Model (M) Comp. time (OP) Comp. time (OP seq.) OPseq./OP SVP2020/OP SVP2004/OP SVP2020/OP seq. SVP2004/OP seq.
1.5 1h 18mn 31s 5h 31mn 54s 4.23 0.89 0.79 0.21 0.19
RMT{ 2.0 1h 11mn 4s 3h 36mn 42s 3.0 0.92 0.87 0.30 0.28
2.5 0h 44mn 23s 2h 52mn 52s 3.89 0.76 0.19

1.5 14h 1mn 41s 0.11 0.11
Wind{ 2.0 4d 4h 10mn 29s 0.70 0.92
2.5 8d 1h 45mn 10s 1.03

Notes. Likewise, the last two columns show the ratios of the SVP runs to the sequential OP calculations. ‘Type’ refers to the kind of model (‘RMT’ and ‘Wind’ for RMT mixing and wind models, respectively). As wind model evolutions are very long, no sequential OP run was attempted. The SVP2020 wind model could not be evolved to the end of the MS and has been ignored.

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.