公开(公告)号：US11216209B2

公开(公告)日：2022-01-04

申请号：US16365080

申请日：2019-03-26

Applicant: Western Digital Technologies, Inc.

Inventor： William Bernard Boyle

IPC: H04L9/32 ,  G06F3/06 ,  G06F21/34 ,  G06F13/40 ,  G06F21/60

Abstract: A storage device comprises a storage medium, storage controller, a host interface, and a bridge slot. The storage controller is configured to control read and write operations to the storage medium and operates according to a firmware written by a storage device manufacturer. The bridge slot is configured to receive a removable bridge storing software written by a third-party different from the storage device manufacturer. The removable bridge is configured to intercept a first command sent from the host system to the storage controller, modify the first command according to the software stored on removable bridge, and transmit the first command to the storage controller.

公开(公告)号：US20240338312A1

公开(公告)日：2024-10-10

申请号：US18232155

申请日：2023-08-09

Applicant: Western Digital Technologies, Inc.

Inventor： Daniel Joseph Linnen ,  William Bernard Boyle ,  Ariel Navon ,  Shay Benisty ,  Alexander Bazarsky

IPC: G06F12/02 ,  G06N3/0455

CPC classification number: G06F12/0246 ,  G06N3/0455

Abstract: Data storage devices configured to exploit generative-adversarial-networks (GANs) are described herein, including super-resolution GANs (SRGANs). In some examples, a GAN-based decoding (reconstruction) procedure is implemented within a data storage controller to replace or supplement an error correction coding (ECC) decoding procedure to permit a reduction in the number of parity bits used while storing the data. In other examples, soft bit information is exploited using GANs during decoding. A dissimilarity matrix may be generated to represent differences between an initial image and a GAN-reconstructed image, with matrix values mapped into low-density parity check (LDPC) codewords to facilitate LDPC decoding of data. In still other examples, confidence information obtained from a GAN is incorporated into image pixels. In some examples, GAN reconstruction of data is limited to modifying valley bits. Multiple GANs may be used in parallel with their outcome aggregated. System and method examples are provided.

公开(公告)号：US10572407B2

公开(公告)日：2020-02-25

申请号：US15675018

申请日：2017-08-11

Applicant: Western Digital Technologies, Inc.

Inventor： Jun Xu ,  Junpeng Niu ,  William Bernard Boyle ,  Jie Yu

IPC: G06F3/06 ,  G06F13/16

Abstract: A data storage system may include one or more storage arrays. Each storage array may include a first set of solid-state drives and a first set of striped hard disk drives. Each solid-state drive of the first set of solid-state drives has a first data throughput and the set of stripe hard disk drives has a second data throughput. The second data throughput of the first set of striped hard disk drives is within a threshold throughput of the first data throughput. The data storage system also includes a processing device configured to receive an access request to write first data to the storage array and determine a read access frequency of the first data. The processing device may also be configured to determine a write access frequency of the first data and write the first data to the first set of solid-state drives or the first set of striped hard disk drives, based on the read access frequency and the write access frequency.

公开(公告)号：US20240338311A1

公开(公告)日：2024-10-10

申请号：US18233125

申请日：2023-08-11

Applicant: Western Digital Technologies, Inc.

Inventor： Ariel Navon ,  Shay Benisty ,  Alexander Bazarsky ,  Daniel Joseph Linnen ,  William Bernard Boyle

IPC: G06F12/02 ,  H03M13/11

CPC classification number: G06F12/0238 ,  H03M13/1105

Abstract: Data storage devices configured to exploit generative-adversarial-networks (GANs) are described herein, including super-resolution GANs (SRGANs). In some examples, a GAN-based decoding (reconstruction) procedure is implemented within a data storage controller to replace or supplement an error correction coding (ECC) decoding procedure to permit a reduction in the number of parity bits used while storing the data. In other examples, soft bit information is exploited using GANs during decoding. A dissimilarity matrix may be generated to represent differences between an initial image and a GAN-reconstructed image, with matrix values mapped into low-density parity check (LDPC) codewords to facilitate LDPC decoding of data. In still other examples, confidence information obtained from a GAN is incorporated into image pixels. In some examples, GAN reconstruction of image data is limited to modifying valley bits. Multiple GANs may be used in parallel with their outcome aggregated. System and method examples are provided.

公开(公告)号：US20240338275A1

公开(公告)日：2024-10-10

申请号：US18232105

申请日：2023-08-09

Applicant: Western Digital Technologies, Inc.

Inventor： Ariel Navon ,  Shay Benisty ,  Alexander Bazarsky ,  Daniel Joseph Linnen ,  William Bernard Boyle

IPC: G06F11/10

CPC classification number: G06F11/1068 ,  G06F11/1048

Abstract: Data storage devices configured to exploit generative-adversarial-networks (GANs) are described herein, including super-resolution GANs (SRGANs). In some examples, a GAN-based decoding (reconstruction) procedure is implemented within a data storage controller to replace or supplement an error correction coding (ECC) decoding procedure to permit a reduction in the number of parity bits used while storing the data. In other examples, soft bit information is exploited using GANs during decoding. A dissimilarity matrix may be generated to represent differences between an initial image and a GAN-reconstructed image, with matrix values mapped into low-density parity check (LDPC) codewords to facilitate LDPC decoding of data. In still other examples, confidence information obtained from a GAN is incorporated into image pixels. In some examples, GAN reconstruction of data is limited to modifying valley bits. Multiple GANs may be used in parallel with their outcome aggregated. System and method examples are provided.