Leveraging Large-Scale TLC 3D NAND Flash Characterization to Investigate Early Lifetime Reliability of SSDs through Extreme Value Statistics
Solid State Drives based on the Triple Level Cell 3D NAND Flash technology are becoming the state-of-the-art in storage architecture. Their reliability analysis is a must. This work presents: i) a large-scale statistical characterization of the errors in the codewords of a TLC 3D NAND Flash Solid State Drive and ii) provides an extreme value statistics model based on the Points Over Threshold approach to capture the early lifetime variations. The model provides a valuable tool for storage designers to implement reliability management strategies since the early drive deployment in the field.
Authors
- Lorenzo Zuolo (ORCID: https://orcid.org/0000-0002-0625-0540)
- Luca Crippa (ORCID: https://orcid.org/0000-0003-1492-9542)
- Rino Micheloni (ORCID: https://orcid.org/0000-0002-3400-2624)
- Alessia Marelli (ORCID: https://orcid.org/0000-0002-2789-4281)
- P. Olivo (ORCID: https://orcid.org/0000-0002-8751-4666)
- Cristian Zambelli (ORCID: https://orcid.org/0000-0001-8755-0504)
Institutions
- University of Ferrara (IT)
Publication Details
- Published
- 2025-03-30
- DOI
- https://doi.org/10.1109/irps48204.2025.10983168
- Primary Topic
- Advanced Data Storage Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00