Abstract
This publication discloses systems and methods for enhancing computing hardware reliability through repairability-driven error detection. Upper-layer software components select a minimum repair action to address hardware or low-level software errors. Each error type is associated with a tagged minimum repair action, which may be determined statically or dynamically at runtime. Machine learning models or expert inputs can be utilized to derive the tagged repair actions based on historical error logs and remediation probabilities. Additional environmental or diagnostic conditions may guide the selection of a specific repair action. The system evaluates both the execution time and the diagnostic time to select actions that optimize recovery speed.
Keywords: repairability, error detection, hardware accelerators, machine learning, fault diagnosing, deep learning, runtime mitigation, troubleshooting, probability.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
N/A, "Systems and Methods for Repairability-Driven Error Detection and Mitigation", Technical Disclosure Commons, (April 01, 2026)
https://www.tdcommons.org/dpubs_series/9690