Delphi.science

FP64 Performance and the Limits of Ozaki Emulation on HPC Systems

Source: arxiv.org/abs/2611.13332

Katarzyna, Swirydowicz, Nicholas Malaya, Noah Wolfe, Paul Bauman, Pieter Ghysels, Jonathan Belof

arXivarxiv/2611.13332Nov 9, 2026cs.DCscore 0

A growing industry narrative holds that native double precision can be deprioritized in HPC accelerators, on the hypothesis that low-precision tensor cores can emulate FP64 on demand, and that most HPC workloads are bandwidth-bound. This paper evaluates some of those claims and explores the importance of FP64 in scientific computing. We show roofline profiling data of representative HPC workloads, across a variety of common computational motifs, to evaluate the arithmetic-intensity and accuracy requirements of these workloads. We also consider FP64 emulation via the Ozaki method to ascertain…

JSONText

Comments and reviews

No comments yet.

Supports Markdown, links, code fences, inline LaTeX ($x$ or \(x\)), and display LaTeX ($$...$$ or \[...\]).

Drafts stay in this tab.

API

curl /api/v1/records/arxiv/2611.13332
curl -X POST /api/v1/records/arxiv/2611.13332/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.13332/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'