perf: cap PyTorch threads, bump fix_illumination forks, skip intermediate pyramids#113
Open
FIrgolitsch wants to merge 6 commits into
Open
perf: cap PyTorch threads, bump fix_illumination forks, skip intermediate pyramids#113FIrgolitsch wants to merge 6 commits into
FIrgolitsch wants to merge 6 commits into
Conversation
This was referenced Apr 30, 2026
FIrgolitsch
force-pushed
the
pr-m-gpu-kvikio
branch
from
7544bf5 to
95f24bc
Compare
FIrgolitsch
force-pushed
the
pr-m-gpu-kvikio
branch
from
95f24bc to
14bfd65
Compare
FIrgolitsch
force-pushed
the
pr-m-gpu-kvikio
branch
from
14bfd65 to
ab73c15
Compare
FIrgolitsch
force-pushed
the
pr-m-gpu-kvikio
branch
from
ab73c15 to
a6f274b
Compare
Squash of resample/GPU/workflow work on dev: - Round-robin GPU assignment for resample_mosaic_grid (linumpy.gpu) - Per-gpu active-slot counters via Helpers.gpuPinBlock for fair maxForks balance - CUDA device pin in prefetch worker thread + workflow CUDA_VISIBLE_DEVICES - Keeps simple per-tile CPU-read -> GPU-rescale -> CPU-write path
Heavy downsampling of small axes (e.g. Z=5 by factor 0.1) used to round to zero, producing zero-sized output and ZeroDivisionError downstream. Clamp output_shape to a minimum of 1 in both rescale() and out_tile_shape.
FIrgolitsch
force-pushed
the
pr-m-gpu-kvikio
branch
from
a6f274b to
f6c23e5
Compare
This was referenced May 20, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR — Pipeline performance tuning
Three measurement-driven perf improvements to the
reconst_3dNextflow pipeline:fix_illuminationthread cap.linum_fix_illumination_3d.pynow callsconfigure_all_libraries()to cap PyTorch threads, preventing oversubscription when running multiple forks.fix_illuminationmaxForks 1 → 4. Measured: BaSiCPy/PyTorch uses ~374 MiB per fork; 4 forks ≈ 1.5 GB, well under per-GPU capacity.--n_levels 0on intermediate steps avoids wasted multiscale generation that's only needed on final outputs.