Eyal Nitzan | Senior Staff Engineer, Machine Learning R&D
KIOXIA

Future of Memory and Storage - Day 1 @ 08:35

DNN-based Read Flow for NAND Flash Memory Controllers

Soft-input reads in NAND flash memory require multiple read operations, creating significant performance overhead. Consequently, NAND flash memory devices usually operate with limited number of read operations and various types of DSP techniques are applied to maintain high read accuracy. In this work, we introduce a deep neural network (DNN)-based read flow that enhances read performance under limited number of reads by dynamically adapting read thresholds on a per–row basis. Unlike conventional controllers that apply uniform thresholds across an entire NAND block, the proposed system uses a compact DNN estimator to determine optimal thresholds for each row in real time. This reduces read-retry rates, improves throughput, and operates efficiently in streaming mode. A lightweight threshold table stores per-block indices, minimizing metadata overhead while enabling fast per-row threshold selection. Additionally, the system employs a small number of optimized “mock reads” that provide highly accurate threshold estimation even under severe device stress. The proposed DNN-based read flow offers a scalable, efficient solution for next-generation NAND flash controllers.

Eyal Nitzan, Senior Staff Engineer, Machine Learning R&D, KIOXIA

×

Eyal Nitzan / DNN-based Read Flow for NAND Flash Memory Controllers

Eyal Nitzan, Senior Staff Engineer, Machine Learning R&D,
KIOXIA

Eyal Nitzan is a Senior Staff Algorithm Engineer in Machine Learning R&D at Kioxia Corporation, where he develops machine-learning and signal-processing algorithms for NAND flash memory controllers, focusing on adaptive read optimization and threshold estimation. He is a co-inventor on multiple ...

CloseMore