sensitivity analysis

[Ating24]

When I was training a polynomial chaotic proxy model using an old version of MOOSE, I used 120 CPUs to calculate one example, but in the end, only one CPU was allocated to this example for calculation

[grmnptr]

Please update your moose, managing the number of processors per app was added in 2021 to the stochastic tools module. Originally, the batch-reset/restore options used one process per app.

[Ating24]

Is there any other way to set a single instance CPU on the old MOOSE? Does having too many polynomial chaotic trainers affect computational efficiency?

[Ating24]

For example, modifying the batch-reset/restore ?

[Ating24]

Can we use code modification to make batch reset/restore options use multiple CPUs?