bench(datapipes): mesh memmap vs zarr benchmark suite#1829
Draft
negin513 wants to merge 1 commit into
Draft
Conversation
Split out of NVIDIA#1798 per review: the reader-level (self-contained prototypes incl. the pmsh-soup layout control), recipe-pipeline, and per-stage profiling scripts behind that PR's performance numbers. benchmark_mesh_vs_zarr.py runs standalone on main; the recipe-pipeline and per-stage scripts additionally need the zarr dataset configs from the recipe follow-up PR at runtime. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
5 tasks
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Split out of #1798 per review: the benchmark harness behind that PR's performance numbers (~1,060 lines).
What's included
benchmark_mesh_vs_zarr.py: reader-level benchmark —.pmsh/memmapMeshReadervs self-contained zarr prototypes (zarr-python and tensorstore), synthetic and real (DrivAerML) modes, cold/warm passes, multi-worker scaling. Includes thepmsh souplayout-control store that separates soup-relayout effects from storage-format effects — the control the review asked to see numbers for.benchmark_recipe_pipeline_mesh_vs_zarr.py: end-to-end recipe DataLoader comparison (benchmark_io).profile_recipe_stages.py: per-stage (read / transform / collate) profiling.Dependencies
benchmark_mesh_vs_zarr.pyis self-contained and runs onmaintoday (its zarr readers are the prototypes that predate and motivated the library'sZarrMeshReader). The recipe-pipeline and per-stage scripts load the zarr dataset configs from #1828 at runtime, so the full suite needs #1798 → #1828 merged first. Draft until then.Checklist
🤖 Generated with Claude Code