Sentinel Cells Enabled Fast Read for NAND Flash


Qiao Li, Department of Computer Science, City University of Hong Kong; Min Ye, YEESTOR Microelectronics Co., Ltd; Yufei Cui, Department of Computer Science, City University of Hong Kong; Liang Shi, School of Computer Science and Software Engineering, East China Normal University; Xiaoqiang Li, YEESTOR Microelectronics Co., Ltd; Chun Jason Xue, Department of Computer Science, City University of Hong Kong


With latest development, NAND flash is experiencing increased errors. The read reference voltages are the key factor for RBER seen by ECC. The limited error correction capability of ECC determines a value range that the read voltages should fall into, otherwise a read failure followed by a read retry with tuned read voltage, would happen. Therefore, finding a correct read voltage with the smallest number of read failures has been a hot research problem. Previous methods in the literature are designed to either progressively tune the voltage value or empirically predict a read voltage based on error models. However, straightforward tuning leads to unpredictable large number of read retries, whereas complex modeling brings large overhead. This paper proposes a novel approach, by reserving a small set of cells as sentinels, which directly tell us the optimal voltage, as drifting caused errors exhibits strong locality. Experiments demonstrate the proposed technique is both efficient and effective.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

@inproceedings {234749,
author = {Qiao Li and Min Ye and Yufei Cui and Liang Shi and Xiaoqiang Li and Chun Jason Xue},
title = {Sentinel Cells Enabled Fast Read for {NAND} Flash},
booktitle = {11th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 19)},
year = {2019},
address = {Renton, WA},
url = {},
publisher = {USENIX Association},
month = jul,