scripts/terrain: pack tiler worklist into a numpy int64 array#470
Merged
Conversation
The worklist was a Python set of (z,x,y) tuples plus a redundant per_level dict of (x,y) tuples (~370 bytes/tile), so a 12.5M-tile z12 build held ~4.6GB in the parent and got OOM-killed on a small box. ProcessPoolExecutor.map also submitted the whole list eagerly, re-materialising it as futures. Pack each tile into one int64 (z<<40|x<<20|y) in a numpy array, dedup+sort with np.unique (same lexicographic z,x,y order as the old tuple sort), build per_level availability only when writing layer.json, and feed the pool in 100k-tile batches so the parent never materialises all futures at once. A 12.5M worklist is now ~100MB. Output is unchanged: a 2x2deg z10-12 build matches the old code - layer.json byte-identical and all 2870 tiles identical when decompressed.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The worklist was a Python set of (z,x,y) tuples plus a redundant per_level dict of (x,y) tuples (~370 bytes/tile), so a 12.5M-tile z12 build held ~4.6GB in the parent and got OOM-killed on a small box. ProcessPoolExecutor.map also submitted the whole list eagerly, re-materialising it as futures.
Pack each tile into one int64 (z<<40|x<<20|y) in a numpy array, dedup+sort with np.unique (same lexicographic z,x,y order as the old tuple sort), build per_level availability only when writing layer.json, and feed the pool in 100k-tile batches so the parent never materialises all futures at once. A 12.5M worklist is now ~100MB. Output is unchanged: a 2x2deg z10-12 build matches the old code - layer.json byte-identical and all 2870 tiles identical when decompressed.