Hugging Face datasets gets major streaming shuffle speedup via optimized Arrow C++
lhoestq · x · 2026-09-07
Quentin Lhoest, author of Hugging Face's datasets library, credits a Google Cloud Storage team contribution for a big speedup when streaming datasets with approximate shuffling. Arrow batches are now buffered, concatenated, and row-shuffled entirely using optimized Arrow C++ operations instead of slower Python-side handling.
Related event: Hugging Face datasets gets major streaming shuffle speedup(2 posts)→
More from Infra
- AWS on agent prompt caching: explicit + implicit patterns to slash token costs — AWS ML Blog · 2026-10-27
- Thieves spoofing GPS steal $100M+ in data center cargo since June — Ars Technica AI · 2026-10-27
- Lam Research: weekly token usage up ~250x in 18 months — Beth_Kindig · 2026-10-27
- AWS shows how to build enterprise agentic retrieval on Bedrock with CDK — AWS ML Blog · 2026-10-27
- Encrypted GPT-2 speedrun challenges FHE inference: best is 3.031 s/token, goal sub-1s — tensorqt · 2026-10-27
- Open-source Latch compiles PyTorch into training megakernels, 2.75x faster on H100 — sguada · 2026-10-27