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.

Sign up for GitHub

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

Jump to bottom

Decision about extending OpenMP or MPI first #443

Closed
MarionBWeinzierl opened this issue Nov 20, 2023 · 4 comments
Closed

Decision about extending OpenMP or MPI first #443

MarionBWeinzierl opened this issue Nov 20, 2023 · 4 comments
Assignees
@draenog @einola @TomMelt @MarionBWeinzierl @timspainNERSC
Labels
ICCS Tasks or reviews for the ICCS team question Further information is requested
Milestone
3 Stand-alone model

Comments

@MarionBWeinzierl
Copy link
Contributor

MarionBWeinzierl commented Nov 20, 2023 •
edited
Loading

Currently, there are OpenMP directives in the dynamics code, and an MPI implementation is being added to the thermodynamics code in #331 .

The question is now whether to go forward with #120 or with #334 first to achieve a hybrid implementation.

The former will probably be quicker.

@einola has set the 1.1.2024 as a deadline for a simple MPI implementation which will allow coupling.
@MarionBWeinzierl MarionBWeinzierl added question Further information is requested ICCS Tasks or reviews for the ICCS team labels Nov 20, 2023
@MarionBWeinzierl MarionBWeinzierl added this to the 3 Stand-alone model milestone Nov 20, 2023
@timspainNERSC
Copy link
Collaborator

@einola has set the 1.1.2004 as a deadline for a simple MPI implementation which will allow coupling.

For clarity, 1.1.2024, else we are very late!

@MarionBWeinzierl
Copy link
Contributor Author

@einola has set the 1.1.2004 as a deadline for a simple MPI implementation which will allow coupling.

For clarity, 1.1.2024, else we are very late!

Sorry! :D I have corrected the date.

@einola
Copy link
Member

einola commented Nov 23, 2023

If you say that #120 would be quicker, let's do that first.

We know that we need MPI, and having some MPI parallelisation of the dynamics is nice, but we're not entirely sure if hybrid MPI+OpenMP is faster than only MPI.

@TomMelt
Copy link
Contributor

TomMelt commented Jul 22, 2024

we did MPI first. OpenMP will be added in near future

@TomMelt TomMelt closed this as completed Jul 22, 2024
@github-project-automation github-project-automation bot moved this from Todo to Done in neXtSIM_DG overview Jul 22, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@draenog draenog

@einola einola

@TomMelt TomMelt

@MarionBWeinzierl MarionBWeinzierl

@timspainNERSC timspainNERSC

Labels
ICCS Tasks or reviews for the ICCS team question Further information is requested
Projects
neXtSIM_DG overview
Status: Done
Milestone
3 Stand-alone model
Development

No branches or pull requests
5 participants
@draenog @einola @TomMelt @MarionBWeinzierl @timspainNERSC