Skip to content

Update checkpointing docs to include new functionality #227

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

Open
andrewcoughtrie wants to merge 1 commit into
base: main
Choose a base branch
from
+13 −5
Open

Update checkpointing docs to include new functionality #227

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 13 additions & 5 deletions documentation/source/user_guide/checkpointing/lfric_atm_checkpoint.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,18 +9,26 @@
LFRic atmosphere checkpoint/restart system
==========================================

The LFRic atmosphere ``lfric_atm`` application can be configured to generate a
checkpoint dump at the end of each model run. The checkpoint dump can be read in
The LFRic atmosphere ``lfric_atm`` application can be configured to generate
checkpoint dumps at multiple points within a model run. The checkpoint dump can be read in
Copy link
Collaborator

@stevemullerworth stevemullerworth Feb 5, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
checkpoint dumps at multiple points within a model run. The checkpoint dump can be read in
checkpoint dumps at multiple points within a model run. A checkpoint dump can be read in

by a new integration of the model allowing further timesteps to be run. The dump
Copy link
Collaborator

@stevemullerworth stevemullerworth Feb 5, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
by a new integration of the model allowing further timesteps to be run. The dump
by a new integration of the model allowing further timesteps to be run. Each dump

is written using XIOS.

Requesting checkpoint restart
-----------------------------

Set ``checkpoint_write=.true.`` in the ``io`` namelist of the model
configuration to generate a checkpoint dump at the end of a model run. The
checkpoint dump will be named after the ``checkpoint_stem_name`` string in the
``files`` namelist appended with the number of the last timestep of the run.
configuration to turn on checkpoint writing. The times when checkpoint files
will be written are defined by the ``checkpoint_times`` list. The
``checkpoint_times`` must be real values in seconds corresponding to an integer
number of timesteps in the model run (e.g. if ``dt=0.5`` the
``checkpoint_times`` can include ``0.5`` and ``1.0`` etc. but **not** ``0.3``).
Setting ``end_of_run_checkpoint=.true.`` will write a checkpoint file at the end
of the run. Only one checkpoint file will be written at the end of a run even if
both the final timestep is included in the ``checkpoint_times`` list and the
``end_of_run_checkpoint`` flag is set to ``.true.``. The checkpoint dumps will
be named after the ``checkpoint_stem_name`` string in the ``files`` namelist
appended with timestep number of the checkpoint time.

Set ``checkpoint_read=.true.`` in the ``io`` namelist to restart a run from an
existing checkpoint dump. The expected start timestep will be defined by
Expand Down
Loading