IP Library Granted Patent US 12,619,533
Granted Patent B1
US 12,619,533 · App. 19/035,262 · Granted May 5, 2026

Data storage device and method for non-balanced mapping for variable read resolution

Inventors: Eran Sharon (Rishon Lezion, IL); Ariel Navon (Revava, IL)
Assignee: Sandisk Technologies, Inc.
G06F12/0246G06F2212/7201
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,619,533
App. No.
19/035,262
Filed
Jan 23, 2025
Granted
May 5, 2026
Kind
B1
Art Unit
2132
USPC
711/103
Abstract

Data is often stored in a multi-level memory (e.g., a quad-level cell (QLC)) using a mapping that balances the bit error rate among the various pages in the memory. The number of sense operations needed to read each page is also relatively balanced. The data storage device presented herein can decide to store data using a non-balanced mapping if there is a desire to later read a low-resolution version of the data. With a non-balanced mapping, fewer sense operations are needed to read the lower page of data, which contains the low-resolution version of the data. Other pages can be read if a higher-resolution version of the data is desired, and techniques can be used to mitigate the higher bit error rate that may be encountered in those other pages because of use of the non-balanced mapping.

Claims (33)

1 . A data storage device comprising:

a memory comprising multi-level memory cells; and

one or more processors, individually or in combination, configured to:

determine whether there is a need for providing a low-resolution version of data;

in response to determining that there is not a need for providing a low-resolution version of the data, store the data in the memory using a first mapping; and

in response to determining that there is a need for providing a low-resolution version of the data, store the data in the memory using a second mapping, wherein reading only a lower page of the data provides the low-resolution version of the data, and wherein fewer memory sense operations are needed to read the lower page when the data is stored using the second mapping than when the data is stored using the first mapping.

2 . The data storage device of claim 1 , wherein the first mapping comprises a 4-4-3-4 mapping and the second mapping comprises a 1-2-4-8 mapping.

3 . The data storage device of claim 1 , wherein the first mapping comprises a bit error rate (BER) balanced mapping and the second mapping comprises a non-BER-balanced mapping.

4 . The data storage device of claim 1 , wherein the determining is performed in response to the data being received from a host.

5 . The data storage device of claim 1 , wherein the determining is performed in response to re-mapping the data, which was previously stored in the memory using a different mapping.

6 . The data storage device of claim 1 , wherein most-significant bits of the data are stored in the lower page.

7 . The data storage device of claim 1 , wherein the low-resolution version of the data comprises a low-resolution version of a parameter of a machine learning model.

8 . The data storage device of claim 1 , wherein the low-resolution version of the data comprises a low-resolution version of an image.

9 . The data storage device of claim 1 , wherein the memory comprises a three-dimensional memory.

10 . In a data storage device comprising multi-level memory cells, a method comprising:

determining whether data is designated for reading with dynamic read resolution; and

in response to determining that the data is designated for reading with dynamic read resolution, storing the data in the memory using a non-BER (bit-error-rate) balanced mapping, wherein fewer memory sense operations are needed to read page(s) of the data for a low-resolution read when the data is stored using the non-BER-balanced mapping instead of a BER-balanced mapping.

11 . The method of claim 10 , further comprising:

performing an operation to mitigate non-balanced BER caused by using the non-BER-balanced mapping.

12 . The method of claim 11 , further comprising:

reading a higher-resolution of the data by reading more pages of the data from the memory.

13 . The method of claim 10 , wherein the data comprises an image.

14 . The method of claim 10 , wherein the data comprises a parameter of a machine-learning model.

15 . The method of claim 10 , wherein the determining is performed in response to the data being received from a host.

16 . The method of claim 10 , wherein the determining is performed in response re-mapping the data, which was previously stored in the memory using a different mapping.

17 . The method of claim 10 , wherein the data is explicitly designated for reading with dynamic read resolution by a host.

18 . The method of claim 10 , wherein the data is implicitly designated for reading with dynamic read resolution by a host.

19 . The method of claim 10 , wherein the BER balanced mapping comprises a 4-4-3-4 mapping and the non-BER balanced mapping comprises a 1-2-4-8 mapping.

20 . A data storage device comprising:

a memory comprising multi-level memory cells; and

means for:

selecting a mapping from a plurality of mappings based on whether the data is designated for variable read resolution; and

writing the data in the memory using the selected mapping.

Assignments (2)
SECURITY AGREEMENT Recorded Apr 25, 2025
From: SANDISK TECHNOLOGIES, INC.
To: JPMORGAN CHASE BANK, N.A., AS COLLATERAL AGENT
Reel/Frame 071050/0001 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 23, 2025
From: SHARON, ERAN; NAVON, ARIEL
To: SANDISK TECHNOLOGIES, INC.
Reel/Frame 069981/0706 →
References Cited (22)
US 8996793B1 · Steiner · 2015 [cited by examiner]
US 9395927B2 · Pekny · 2016 [cited by applicant]
US 11853572B2 · Gopalakrishnan et al. · 2023 [cited by applicant]
US 20140013033A1 · Sharon · 2014 [cited by examiner]
US 20140157086A1 · Sharon · 2014 [cited by examiner]
US 20150378801A1 · Navon · 2015 [cited by examiner]
US 20210132818A1 · Hallak et al. · 2021 [cited by applicant]
KR 101206820 · 2012 [cited by applicant]
U.S. Appl. No. 19/035,315, filed Jan. 23, 2025 entitled “Data Storage Device and Method for Bit Error Rate (BER) Balancing Based on Dynamic Sense Time.” [cited by applicant]
U.S. Appl. No. 18/945,932, filed Nov. 13, 2024 entitled “Data Storage Device and Method for Direct Data Organization in Multi-Level-Cell Memory.” [cited by applicant]
U.S. Appl. No. 18/230,576, filed Aug. 4, 2023 entitled “Computational Storage Device with Computational Precision-Defined Fixed Point Data Grouping and Storage Management.” [cited by applicant]
Courbariaux, Matthieu et al.; “Binarized Neural Networks: Training Neural Networks with Weights and Activations Constrained to +1 or <1”; downloaded from the Internet on Jan. 23, 2025 at https://arxiv.org/abs/1602.02830… [cited by applicant]
Jiang, Chutian; “Efficient Quantization Techniques for Deep Neural Networks”; downloaded from the Internet on Jan. 23, 2025 at https://ieeexplore.ieee.org/document/9707043; 2021 International Conference on Signal Proces… [cited by applicant]
Mishra, Rahul et al.; “A Survey on Deep Neural Network Compression: Challenges, Overview, and Solutions”; downloaded from the Internet on Jan. 23, 2025 at https://arxiv.org/abs/2010.03954; arXiv:2010.03954; Oct. 5, 2020… [cited by applicant]
Peng, Fu et al.; “Training Quantized Deep Neural Networks via Cooperative Coevolution”; downloaded from the Internet on Jan. 23, 2025 at https://arxiv.org/pdf/2112.14834; arXiv:2112.14834v3; InInternational Conference o… [cited by applicant]
“Quantization”; Guidance documentation webpage; downloaded from the Internet on Jan. 23, 2025 at https://pytorch.org/docs/stable/quantization.html; The Linux Foundation, PyTorch Contributors; 2023; 15 pages. [cited by applicant]
Brookes, Tim; “What Is Dynamic Resolution Scaling (DRS)?”; downloaded from the Internet on Nov. 13, 2024 at https://www.howtogeek.com/764408/what-is-dynamic-resolution-scaling-drs/; Nov. 14, 2021; How-To Geek; Valnet Pu… [cited by applicant]
Ponti, M. et al.; “Compact Color Features with Bitwise Quantization and Reduced Resolution for Mobile Processing”; downloaded from the Internet on Nov. 13, 2024 at Ponti MobileImagingSIP2013.pdf; InProceedings 2013; 4 p… [cited by applicant]
“TensorFlow: An end-to-end platform for machine learning”; product page; downloaded from the Internet on Nov. 13, 2024 at https://www.tensorflow.org; Oct. 28, 2024; 10 pages. [cited by applicant]
Wang, Cheng et al.; “Channel Modeling and Quantization Design for 3D NAND Flash Memory”; Entropy, vol. 25; Jun. 21, 2023; 17 pages. [cited by applicant]
Non-final Office Action mailed Oct. 1, 2025 for U.S. Appl. No. 18/945,932. [cited by applicant]
Yue, J. et al; “Reducing Read Latency in MLC PCM”; In2016 IEEE International Conference on Networking, Architecture and Storage (NAS); IEEE; Aug. 8, 2016; 6 pages. [cited by applicant]