Archivio Istituzionale della RicercaNowadays, AI applications are becoming extremely popular in our everyday life as well as for the industry. Recent incidents involving hyperscalers have revealed that even cloudbased datacenter hardware can experience failures leading to Silent Data Corruptions (SDCs), also called Silent Data Errors (SDEs). This Special Session delves into the implications of such failures on AI workloads, both during training and inference, and explores methodologies for efficiently detecting SDCs or SDEs through dedicated monitoring phases.
Special Session: Trustworthy Hardware-AI at the Cloud / Angione, Francesco; Bernardi, Paolo; Bosio, Alberto; Dattatraya Dixit, Harish; Pappalardo, Salvatore; Ruospo, Annachiara; Sanchez, Ernesto; Sinha, Arani; Turco, Vittorio. - ELETTRONICO. - (2025). (Intervento presentato al convegno IEEE VLSI Test Symposium 2025 tenutosi a Tempe, Arizona (USA) nel 28-30 April 2025) [10.1109/VTS65138.2025.11022869].
Special Session: Trustworthy Hardware-AI at the Cloud
Francesco Angione;Paolo Bernardi;Annachiara Ruospo;Ernesto Sanchez;Vittorio Turco
2025
Abstract
Nowadays, AI applications are becoming extremely popular in our everyday life as well as for the industry. Recent incidents involving hyperscalers have revealed that even cloudbased datacenter hardware can experience failures leading to Silent Data Corruptions (SDCs), also called Silent Data Errors (SDEs). This Special Session delves into the implications of such failures on AI workloads, both during training and inference, and explores methodologies for efficiently detecting SDCs or SDEs through dedicated monitoring phases.
| File | Dimensione | Formato | |
|---|---|---|---|
|
VTS25___Special_Session.pdf
accesso aperto
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
Pubblico - Tutti i diritti riservati
Dimensione
643.21 kB
Formato
Adobe PDF
|
643.21 kB | Adobe PDF | Visualizza/Apri |
|
Special_Session_Trustworthy_Hardware-AI_at_the_Cloud.pdf
accesso riservato
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
697.93 kB
Formato
Adobe PDF
|
697.93 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/3000875