Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Trixi displays local, and not global, number of elements / DOFs #1616

Open
benegee opened this issue Aug 29, 2023 · 4 comments · May be fixed by #1865
Open

Trixi displays local, and not global, number of elements / DOFs #1616

benegee opened this issue Aug 29, 2023 · 4 comments · May be fixed by #1865
Labels
low-priority parallelization Related to MPI, threading, tasks etc. question Further information is requested

Comments

@benegee
Copy link
Contributor

benegee commented Aug 29, 2023

When doing parallel computations with Trixi, I noticed that the local number of elements or degrees of freedom, and not the global number, is printed.

For example:

mpiexec() do cmd
  run(`$cmd -n 2 $(Base.julia_cmd()) -e 'using Trixi; trixi_include("Trixi.jl/examples/p4est_2d_dgsem/elixir_euler_sedov.jl")'`)
end

runs a 2D P4est example with 256 elements and 6400 DOFs on 2 processes, and the output is:

[...]
│ SemidiscretizationHyperbolic                                                                     │
[...]
│ total #DOFs: ……………………………………………… 3200                                                             │
[...]
│ P4estMesh{2, Float64}                                                                            │
[...]
│ current #cells: ……………………………………… 128                                                              │

Likewise the analysis callback prints:

[...]
 #DOF:                     3200                alloc'd memory:        343.208 MiB
 #elements:                 128
[...]

I would have expected to see the global number. So, this is just a question whether displaying the local number is intended.

@benegee benegee added question Further information is requested parallelization Related to MPI, threading, tasks etc. low-priority labels Aug 29, 2023
@ranocha
Copy link
Member

ranocha commented Aug 29, 2023

I think both numbers would be interesting.
@sloede?

@sloede
Copy link
Member

sloede commented Aug 30, 2023

I have stumbled across this issue before as well. Thus I agree, it would be great if we could show the global counts instead of the rank 0 counts in the aforementioned locations. Would you be willing to create a PR for this, @bgeihe? I'd be happy to assist in case of questions.

@benegee
Copy link
Contributor Author

benegee commented Aug 30, 2023

Sure!

But to clarify:
Show global numbers instead of local ones?
OR
Show global and local numbers?

@sloede
Copy link
Member

sloede commented Aug 30, 2023

To me, showing only the global numbers makes more sense. I am not sure exactly how much information one can gain from only knowing the local number of elements/DOFs on rank 0 🤷‍♂️

@benegee benegee linked a pull request Mar 7, 2024 that will close this issue
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
low-priority parallelization Related to MPI, threading, tasks etc. question Further information is requested
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants