Data Center Silent Data Errors: Implications to Artificial Intelligence Workloads & Mitigations
Brad C. Bittel, Manu Shamsa, Benson Inkley, Amit Gur, David P. Lerner, M. Adams · 2024
Silent Data Errors (SDEs) are a unique category of errors that result in unpredictable system behavior that is often difficult to detect. SDEs can represent a serious concern to at-scale compute in data center operations. [1], [2] This paper reviews data collected on SDE impacts to Artificial Intelligence (AI) workloads and Intel's SDE mitigation tools available for use in the data center.