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

Update HPC docs with pldd usage #3371

Merged
merged 2 commits into from
Dec 11, 2024
Merged

Update HPC docs with pldd usage #3371

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d6fa877
update hpc with pldd
jessica-mitchell Nov 19, 2024
423ff52
Update doc/htmldoc/hpc/slurm_script.rst
jessica-mitchell Dec 10, 2024
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
25 changes: 17 additions & 8 deletions doc/htmldoc/hpc/slurm_script.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,8 @@ In this example, we are using 1 node, which contains 2 sockets and 64 cores per
export OMP_NUM_THREADS=$SLURM_CPUS_PER_TASK
export OMP_PROC_BIND=TRUE

# Optional
python -c "import nest, subprocess as s, os; s.check_call(['/usr/bin/pldd', str(os.getpid())])" 2>&1 | tee -a "pldd-nest.out"

# On some systems, MPI is run by SLURM
srun --exclusive python3 my_nest_simulation.py
Expand Down Expand Up @@ -174,6 +176,21 @@ will prevent the threads from moving around.

|

::

python -c "import nest, subprocess as s, os; s.check_call(['/usr/bin/pldd', str(os.getpid())])" 2>&1 | tee -a "pldd-nest.out"

Prints out the linked libraries into a file with name ``pldd-nest.out``.
In this way, you can check whether dynamically linked librariries for
the execution of ``nest`` is indeed used. For example, you can check if ``jemalloc`` is used for the network construction
in highly parallel simulations.

.. note::

The above command uses ``pldd`` which is commonly available in Linux distributions. However, you might need to change
the path, which you can find with the command ``which pldd``.

|

You can then tell the job script to schedule your simulation.
Setting the ``exclusive`` option prevents other processes or jobs from doing work on the same node.
Expand Down Expand Up @@ -222,11 +239,3 @@ It should match the number of ``cpus-per-task``.
.. seealso::

:ref:`parallel_computing`








Loading