HAIPO: Hybrid AI Algorithm-Based Post-Fabrication Optimization for Modern 3D NAND Flash Memory

被引:0
作者
Kim, Myungsuk [1 ]
机构
[1] Kyungpook Natl Univ, Sch Comp Sci & Engn, Daegu 37224, South Korea
基金
新加坡国家研究基金会;
关键词
NAND flash memory; optimization; AI; deep learning; evolutionary algorithm; manufacturing process;
D O I
10.3390/pr12122760
中图分类号
TQ [化学工业];
学科分类号
0817 ;
摘要
To successfully meet the various requirements of modern storage systems, NAND flash memory should be highly optimized by precisely tuning a huge number of internal operating parameters. Although 3D NAND flash memory succeeds in increasing the capacity of storage systems, its complex architecture and unique error behavior make such optimization a more difficult and time-consuming process during NAND manufacturing. In this paper, we introduce HAIPO, a novel methodology for post-fabrication optimization of NAND flash memory, which is an essential step in the manufacturing process of modern 3D NAND flash memory to simultaneously meet various requirements on reliability, performance, yield, etc. HAIPO is based on simple machine-learning approaches that consist of (i) a lightweight deep-learning (DL) model to generate initial device parameters and (ii) an evolutionary algorithm (EA) to explore device parameters automatically. To more effectively explore device parameters, we introduce three key guidelines for each generation in the EA: (1) domain-specific rules, (2) recent optimization results, and (3) online Bayesian simulation, respectively, to enable quick optimization for a huge number of device parameters within the limited product turnaround time (TAT). In addition, we integrate two optimization modules with HAIPO to improve optimization efficiency even in environments with severe process variation. We demonstrate the feasibility and effectiveness of HAIPO using real 320 3D TLC/QLC NAND flash chips, showing significant performance and reliability improvements by up to 8.8% and 12% on average, respectively, within a quite limited optimization TAT.
引用
收藏
页数:19
相关论文
共 33 条
  • [1] [Anonymous], 2022, JEDEC JESD218B.02: Solid-State Drive (SSD) Requirements and Endurance Test Method
  • [2] Arrhenius S., 1889, Z. Phys. Chem., V4, P96, DOI DOI 10.1515/ZPCH-1889-0408
  • [3] Bongale P, 2016, IEEE VLSI TEST SYMP
  • [4] Cho J., 2021, P INT SOL STAT CIRC
  • [5] Genetic-algorithm-based method for optimal analog test points selection
    Golonek, T.
    Rutkowski, J.
    [J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2007, 54 (02) : 117 - 121
  • [6] Deep Feature Consistent Variational Autoencoder
    Hou, Xianxu
    Shen, Linlin
    Sun, Ke
    Qiu, Guoping
    [J]. 2017 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2017), 2017, : 1133 - 1141
  • [7] Jang J., 2009, P S VLSI TECHN KYOT
  • [8] Jung S.M., 2006, P INT EL DEV M IEDM
  • [9] A 19 nm 112.8 mm2 64 Gb Multi-Level Flash Memory With 400 Mbit/sec/pin 1.8 V Toggle Mode Interface
    Kanda, Kazushige
    Shibata, Noboru
    Hisada, Toshiki
    Isobe, Katsuaki
    Sato, Manabu
    Shimizu, Yui
    Shimizu, Takahiro
    Sugimoto, Takahiro
    Kobayashi, Tomohiro
    Kanagawa, Naoaki
    Kajitani, Yasuyuki
    Ogawa, Takeshi
    Iwasa, Kiyoaki
    Kojima, Masatsugu
    Suzuki, Toshihiro
    Suzuki, Yuya
    Sakai, Shintaro
    Fujimura, Tomofumi
    Utsunomiya, Yuko
    Hashimoto, Toshifumi
    Kobayashi, Naoki
    Matsumoto, Yuuki
    Inoue, Satoshi
    Suzuki, Yoshinao
    Honda, Yasuhiko
    Kato, Yosuke
    Zaitsu, Shingo
    Chibvongodze, Hardwell
    Watanabe, Mitsuyuki
    Ding, Hong
    Ookuma, Naoki
    Yamashita, Ryuji
    [J]. IEEE JOURNAL OF SOLID-STATE CIRCUITS, 2013, 48 (01) : 159 - 167
  • [10] Smart Electrical Screening Methodology for Channel Hole Defects of 3D Vertical NAND (VNAND) Flash Memory
    Kim, Beomjun
    Seo, Gyeongseob
    Kim, Myungsuk
    [J]. ENG, 2024, 5 (01): : 495 - 512