← 研踪 PaperTrailarXiv:2610.35413 · v1
How Numerical Errors Scale with Hardware
首次提交 2026-10-26 21:25
版本更新 2026-10-26 21:25
雷达观测 2026-10-29 00:36
摘要
In this short technical note, we outline how to bound numerical errors with respect to computational resources. We consider algorithms with known complexity estimates that can be executed in parallel and combine these estimates with parallelization laws. We suppose that the proposed approach generalizes beyond the presented examples and argue that it provides a straightforward and fair framework for comparing algorithms, hardware systems, and software implementations. This is achieved by imposing identical resource or energy budgets on each configuration, and comparing the resulting numerical error evaluations.
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v1 · 2026-10-26 21:25 · 最新观测版本
How Numerical Errors Scale with Hardware
In this short technical note, we outline how to bound numerical errors with respect to computational resources. We consider algorithms with known complexity estimates that can be executed in parallel and combine these estimates with parallelization laws. We suppose that the proposed approach generalizes beyond the presented examples and argue that it provides a straightforward and fair framework for comparing algorithms, hardware systems, and software implementations. This is achieved by imposing identical resource or energy budgets on each configuration, and comparing the resulting numerical error evaluations.
查看此版本解读 →查看此版本原文 ↗