Metraipolis › archive › dc_b0ac01e148be
Calibrating the Checksum: An Empirical False-Positive Bound for One-Sided ABFT on bfloat16 GEMM, and What It Fails to Catch — arXiv:2610.02240
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.02240
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:53:00.187Z
- SHA-256 of the archived text
b148565af73cdd8d581f93f2dba4daab467a8013045eaf7d8b484eaa16cc2acd- Stable URL of this page
https://metraipolis.com/documento/dc_b0ac01e148be
How to cite
Calibrating the Checksum: An Empirical False-Positive Bound for One-Sided ABFT on bfloat16 GEMM, and What It Fails to Catch — arXiv:2610.02240 — arXiv, https://arxiv.org/abs/2610.02240 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b0ac01e148be (Metraipolis archive dc_b0ac01e148be, sha256 b148565af73cdd8d581f93f2dba4daab467a8013045eaf7d8b484eaa16cc2acd)
BibTeX
@misc{dcb0ac01e148be,
title = {Calibrating the Checksum: An Empirical False-Positive Bound for One-Sided ABFT on bfloat16 GEMM, and What It Fails to Catch — arXiv:2610.02240},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.02240},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b0ac01e148be; sha256 b148565af73cdd8d581f93f2dba4daab467a8013045eaf7d8b484eaa16cc2acd},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02240. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Algorithm-based fault tolerance detects corrupted matrix multiplications by comparing a checksum of the product against a checksum recomputed from the factors, flagging a fault when the residual exceeds a threshold. The classical threshold is tau = Ku, derived from a dot-product rounding bound that requires Ku <= 0.01. At bfloat16 with contraction length K = 4096, Ku is approximately 32, so the bound does not hold and the threshold is not defined. Prior work therefore evaluated checksum ABFT only at float32. We supply the missing calibration. This paper is a characterization of a detector under an output-level, post-GEMM perturbation model; it is not a characterization of faults in silicon, and no rate reported here estimates fault incidence in deployed hardware. We fit a Generalized Pareto distribution to 7.99e7 clean per-row residuals and extrapolate a threshold at a per-GEMM false-positive rate of 1e-6, obtaining 3.1445e-6, which sits between the two clean sample maxima we observed. The independence assumption behind the per-GEMM conversion is checked empirically rather than asserted, at a variance ratio of 0.985 against a binomial reference. Against this threshold the detector catches 304 of 400 single exponent-bit flips (76%), failing a bar of 90% that was registered before any data was collected, and catches 0 of 600 low-mantissa flips. Detectability is invariant in K across a 16x range even though the clean noise floor falls as K^-0.4999: signal and floor scale together, so bit position governs detectability and contraction length does not. This falsified the mechanism the study was designed around. We report four registered predictions that failed, including that one, and four retractions of our own earlier findings, two of which were caught by pre-publication audit rather than by prediction. Code, data, and timestamped pre-registration documents are released. Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.DC, nuovi lavori del giorno). Categorie arXiv: Hardware Architecture (cs.AR) ; Distributed, Parallel, and Cluster Computing (cs.DC). Pagina del lavoro: https://arxiv.org/abs/2610.02240 — 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 2026-10-05T22:46:25.547Z dalla pagina https://arxiv.org/list/cs.DC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).