Skip to content

Add some node-load-balancing test coverage #4404

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
roystgnr wants to merge 1 commit into
base: devel
Choose a base branch
from
+20 −3
Open

Add some node-load-balancing test coverage #4404

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
23 changes: 20 additions & 3 deletions examples/adaptivity/adaptivity_ex3/run.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,23 +7,40 @@
example_name=adaptivity_ex3
example_dir=examples/adaptivity/$example_name

# Test with a bunch of different refinement types; while we're at it
# we can get a little test coverage of our nodal partitioning options.
#
# --load-balanced-nodes-bfs doesn't appear to be safe to use on a
# distributed mesh! I see dbg assertion failures from a sweep over as
# few as 15 processors!

run_example "$example_name" refinement_type=h
# Some solvers still give us trouble with too much p
run_example "$example_name" refinement_type=p max_r_steps=4
run_example "$example_name" refinement_type=hp max_r_steps=6
run_example "$example_name" refinement_type=hp max_r_steps=6 --load-balanced-nodes-linear
if [ "x$petscmajor" == "x" ]; then
run_example "$example_name" refinement_type=matchedhp max_r_steps=4 --load-balanced-nodes-petscpartitioner
else
run_example "$example_name" refinement_type=matchedhp max_r_steps=4
fi

# Let's get some 1D coverage too; that's cheap.
run_example "$example_name" dimension=1 refinement_type=h
# Load balancing nodes should be pointless here but we can make sure
# it doesn't regress in this special case.
run_example "$example_name" dimension=1 refinement_type=h --load-balanced-nodes-linear
# After enough p refinement we start breaking non-pivoting ILU preconditioners
run_example "$example_name" dimension=1 refinement_type=p max_r_steps=8
run_example "$example_name" dimension=1 refinement_type=hp max_r_steps=6
run_example "$example_name" dimension=1 refinement_type=matchedhp max_r_steps=4

# And try out another element type
if [ "x$petscmajor" == "x" ]; then
run_example "$example_name" dimension=1 approx_type=HERMITE approx_order=3 refinement_type=h --load-balanced-nodes-petscpartitioner
else
run_example "$example_name" dimension=1 approx_type=HERMITE approx_order=3 refinement_type=h
fi
run_example "$example_name" dimension=1 approx_type=HERMITE approx_order=3 refinement_type=p max_r_steps=6
run_example "$example_name" dimension=1 approx_type=HERMITE approx_order=3 refinement_type=hp max_r_steps=8
run_example "$example_name" dimension=1 approx_type=HERMITE approx_order=3 refinement_type=hp max_r_steps=8 --load-balanced-nodes-linear
run_example "$example_name" dimension=1 approx_type=HERMITE approx_order=3 refinement_type=matchedhp max_r_steps=4

# Examples to use for test coverage
Expand Down