Western Digital Technologies, Inc.

Micron Technology Incorporated

Chao-Han Cheng

Bo Lei

Dongxiang Liao

Mohsen Purahmad

Nian Niles Yang

Zhenlei Shen

Niles Yang

Anubhav Khandelwal

Nan Lu

Lei Zhang

Zelei Guo

Tienchien Kuo

Chung-Yao Pai

Srinivasa Reddy Kunduru

A wake up cadence at which the memory device is to wake up during a low power mode is determined. Based on the wake up cadence, a processing device determines a finite number of pages of the memory device to be scanned per wake up during the low power mode to satisfy a criterion for memory device qualification, the criterion being associated with the memory device's data retention capability. Upon detecting the low power mode, a burst scan operation is performed to scan the finite number of pages in the memory device each time the memory device wakes up at the determined cadence.

Memory sub-system initiated burst scan under low power mode

Apparatus and methods for recovery after an abort event are described. A data storage system may comprise a non-volatile memory device, having one or more wordlines configured to receive a read level voltage, and a controller. The controller is configured to detect whether a write abort event occurred for the data storage system. The controller is configured to determine a first voltage offset based on one or more of a wear-level indication of the non-volatile memory device, or one or more voltage parameters of the non-volatile memory device. The controller is configured to determine, based on the first voltage offset, an adjusted read level voltage. The controller is configured to apply the adjusted read level voltage to a wordline of the non-volatile memory device. The controller is configured to read data, based on the applied adjusted read level voltage, from the wordline of the non-volatile memory device.

Data storage systems and methods for improved recovery after a write abort event

Apparatus, media, methods, and systems for data storage systems and methods for improved recovery after a write abort event are described. A data storage system may comprise a non-volatile memory device, having one or more wordlines configured to receive a read level voltage, and a controller. The controller is configured to detect whether a write abort event occurred for the data storage system. The controller is configured to determine a first voltage offset based on one or more of a wear-level indication of the non-volatile memory device, or one or more voltage parameters of the non-volatile memory device. The controller is configured to determine, based on the first voltage offset, an adjusted read level voltage. The controller is configured to apply the adjusted read level voltage to a wordline of the non-volatile memory device. The controller is configured to read data, based on the applied adjusted read level voltage, from the wordline of the non-volatile memory device.

A data storage system may include a non-volatile memory device, having one or more wordlines configured to receive a read level voltage, and a controller. The controller is configured to detect whether a write abort event occurred for the data storage system. The controller is configured to determine a first voltage offset based on one or more of a wear-level indication of the non-volatile memory device, or one or more voltage parameters of the non-volatile memory device. The controller is configured to determine, based on the first voltage offset, an adjusted read level voltage. The controller is configured to apply the adjusted read level voltage to a wordline of the non-volatile memory device. The controller is configured to read data, based on the applied adjusted read level voltage, from the wordline of the non-volatile memory device. Methods are also described.

Systems, methods, and/or devices are used to manage open blocks within non-volatile storage devices, in order to improve the reliability of non-volatile storage devices. In some embodiments, when a shut-down request is received from a host device, the storage device fetches information about open blocks and their boundary regions susceptible to data reliability issues, and for each identified boundary region, the storage device programs a region contiguous to the identified boundary region. In some embodiments, the device updates an XOR parity table used for XOR parity management with the information that the region contiguous to the identified boundary is programmed. Subsequently, in some embodiments, the storage device can use the information, stored in the contiguous region and/or the XOR parity table, for data recovery in the event of a data loss. As a result, the reliability of the non-volatile storage device is improved.

Method and system for improving open block data reliability

Adaptively changing a fail bit count for an erase operation is disclosed. A memory system may detect an erase stuck bit condition in a group of memory cells. An erase stuck bit condition refers to a situation in which the threshold voltage of at least one memory cell on string tends to stick, such that the string cannot be erased. The memory system performs an action in response to detecting an erase stuck bit condition, in one embodiment. One possible action is to increase a fail bit count for erase operations for other groups of memory cells, which could also potentially suffer from erase bit stuck conditions. This can help reduce erase stress on groups of memory cells. It can also reduce the number of groups of memory cells that need to be retired for failing an erase operation.

Adaptive erase fail bit criteria

A non-volatile storage apparatus includes a set of non-volatile memory cells and one or more control circuits in communication with the set of non-volatile memory cells. The one or more control circuits are configured to identify a most recently written portion of the set of non-volatile memory cells and to compare an error rate of data stored in the most recently written portion with a reference error rate from a reference portion of the set of non-volatile memory cells to determine whether the most recently written portion is fully written or partially written.

Growing community of inventors

San Jose, CA, United States of America

Chao-Han Cheng