Metraipolis › archive › dc_75922f2e771e

You Only Convert Twice: Ozaki Scheme II for Chained Tensor Mode Products — arXiv:2610.05059

Source
arXiv
Original
https://arxiv.org/abs/2610.05059
License (as archived)
CC0-1.0
Archived
2026-10-06T05:29:00.638Z
SHA-256 of the archived text
0db8d3911e4122740e0b19542d444029d5b67cecc67100230dd15839de64dfa5
Stable URL of this page
https://metraipolis.com/documento/dc_75922f2e771e

How to cite

You Only Convert Twice: Ozaki Scheme II for Chained Tensor Mode Products — arXiv:2610.05059 — arXiv, https://arxiv.org/abs/2610.05059 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_75922f2e771e (Metraipolis archive dc_75922f2e771e, sha256 0db8d3911e4122740e0b19542d444029d5b67cecc67100230dd15839de64dfa5)
BibTeX
@misc{dc75922f2e771e,
  title = {You Only Convert Twice: Ozaki Scheme II for Chained Tensor Mode Products — arXiv:2610.05059},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05059},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_75922f2e771e; sha256 0db8d3911e4122740e0b19542d444029d5b67cecc67100230dd15839de64dfa5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05059. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This work proposes OzII-RescaleBE, a method that enables persistent execution of chained tensor mode products in Ozaki scheme II, keeping the intermediate tensors in residue form. By performing rescaling and base extension directly on the residue representation, it requires conversion to residues only at the beginning of the chain and reconstruction only at the end. An implementation using CuTe DSL and INT8 Tensor Cores is provided for verification and evaluation. The method is compared with per-mode composition of Ozaki scheme~II (GEMMul8-Composed) and a dense Kronecker-product formulation (GEMMul8-KRON), both implemented using the GEMMul8 library. Across three datasets with different input distributions, OzII-RescaleBE achieves normwise relative errors ranging from $2.5\times10^{-16}$ to $2.6\times10^{-15}$ for chain depths $d=3$ through $7$. The corresponding errors range from $10^{-16}$ to $10^{-15}$ for GEMMul8-KRON and from $1\times10^{-16}$ to $3\times10^{-16}$ for GEMMul8-Composed and the FP64 chain. At $d=8$, the errors of OzII-RescaleBE increase to between $7\times10^{-14}$ and $6\times10^{-13}$. For mode sizes $n=8$ to $256$, OzII-RescaleBE achieves throughputs ranging from 1.47 to 4.42\,GDoF/s. Its throughput is within 12\,\% of GEMMul8-KRON at $n=8$ and $12$ and exceeds it at larger tested sizes. Compared with GEMMul8-Composed, OzII-RescaleBE achieves $2.1$--$60\times$ the throughput for $n\leq128$, but has 6--12\,\% lower throughput at $n=256$. It also uses less device memory than the GEMMul8 baselines in the measured memory comparisons. These results demonstrate that rescaling and base extension enable accurate residue-domain tensor chains while avoiding repeated intermediate conversions to floating point.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.DC, nuovi lavori del giorno).
Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2610.05059 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T05:24:35.300Z dalla pagina https://arxiv.org/list/cs.DC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).