Skip to content

Swapped cut_through for lineout in cutthrough_timeseries#438

Merged
alhom merged 13 commits intofmihpc:masterfrom
JonasSuni:interpolate_cutthrough_timeseries
Mar 9, 2026
Merged

Swapped cut_through for lineout in cutthrough_timeseries#438
alhom merged 13 commits intofmihpc:masterfrom
JonasSuni:interpolate_cutthrough_timeseries

Conversation

@JonasSuni
Copy link
Contributor

Also fixed some old incompatible code in other files

Testing pending

@JonasSuni
Copy link
Contributor Author

Testing is troublesome, there appears to be a memory leak somewhere

@JonasSuni
Copy link
Contributor Author

python cutthrough_timeseries.py -var "vg_b_vol" -fnr 600 650 -bulkpath '/wrk-vakka/group/spacephysics/vlasiator/3D/FIF/bulk1' -bulkprefix 'bulk1' -point 10 -20 0 20 -20 0 -outputdir "./" -outputname "test.png" -op "x" -re True -npoints 50

It works, but the new npoints parameter should be chosen reasonably. 100 results in memory leak in this case, for instance

ct_test

@JonasSuni JonasSuni marked this pull request as ready for review March 5, 2026 12:48
@JonasSuni
Copy link
Contributor Author

The memory leak: In an interactive job with 32GB memory, the python process is killed after approx 27 timesteps have been processed with npoints=100. In a job with 64GB memory, it is killed after approx 50 timesteps

@JonasSuni
Copy link
Contributor Author

Do we wait for the CI to finish before merging?

@alhom
Copy link
Member

alhom commented Mar 6, 2026

Let me fix the cut-through into this PR and we can have the 0-order work through that, sanely. Also adding some sanity checks for the number of lineout points to warn about undersampling.

alhom and others added 2 commits March 6, 2026 11:48
…used variables to dummy.

Add helper function to find smallest SpatialGrid dx/cell lengths.
@alhom alhom merged commit 4fbe4c9 into fmihpc:master Mar 9, 2026
24 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants