Confusion about the results obtained by specifying parameters for Laghos.

CEED / Laghos

High-order Lagrangian Hydrodynamics Miniapp

http://ceed.exascaleproject.org/miniapps

BSD 2-Clause "Simplified" License

187 stars 60 forks source link

Confusion about the results obtained by specifying parameters for Laghos. #189

Open Wondertolie666 opened 4 months ago

Wondertolie666 commented 4 months ago

When I specified the number of cores, I found that the more cores I specified, the longer the program ran and the worse the results. What is the reason for this. For example, when my - np is 1, the time for each item(CG, Force..) is the shortest and the rate is the highest. At - np 64, the results were actually very poor. Or rather, can the program reflect the performance of the processor and what specific indicators are used to reflect it.

vladotomov commented 4 months ago

Hi @Wondertolie666,

This has not been the case in our tests of Laghos or any of the parallel MFEM examples. You should check your hardware configuration.