Micron Technology Incorporated

Alcatel Lucent

Western Digital Technologies, Inc.

Sandisk Technologies Inc.

Wei Wang

Jiangli Zhu

Ying Yu Tai

Fangfang Zhu

Zhenming Zhou

Frederick Adi

Venkata Naga Lakshman Pasala

Tingjun Xie

Huapeng G Guan

Yipei Yu

Zhenlei Shen

Murong Lang

Seungjune Jeon

Ning Chen

Samir Mittal

Chih-Kuo Kao

Charles See Yeung Kwong

Jason Duong

Zhengang Chen

Xiaoheng Chen

Cheng Yuan Wu

Wu Wang

Jingfeng Yuan

Mei Marker

Yu-Chung Lien

Yang Liu

Paul Roger Stonelake

Li-Te Chang

Kelvin Ross Edmison

Ximin Shan

Hong Yao

Jeffrey L Whaley

Jeffrey Lee Whaley

Nagireddy Chodem

Aspects of the present disclosure configure a system component, such as a memory sub-system controller, to perform adaptive read level threshold voltage operations. The controller determines a first read level offset associated with reading a first set of data from a first level using a first read level of a plurality of read levels. The controller applies the first read level offset to a machine learning model to estimate a second read level offset, associated with reading a second set of data from a second level of the plurality of levels, using a second read level of the plurality of read levels. The controller updates, based on the first read level offset and the estimated second read level offset, a look-up table that includes a set of read level offsets used to read data from the plurality of levels of the individual component.

Model based error avoidance

A first blocks of a set of blocks of a memory device is identified. A die on which the first block resides is identified among a plurality of dies of the memory device. A threshold value associated with the die is selected from a range associated with a projected reliability metric of the die. Responsive to determining that an endurance metric value associated with the die matches the threshold value, a program operation is performed with respect to a second block of the set of blocks.

Adaptive wear leveling for endurance compensation

Methods, systems, and devices for one or more acceleration engines for memory sub-system operations are described. An acceleration engine can perform one or more validation procedures on one or more codewords of a management unit. The acceleration engine can collect validation data for the management unit based on performing the validation procedures. The acceleration engine can aggregate the validation data into group validation data associated with a set of management units. The acceleration engine can transmit the group validation data to firmware of a memory sub-system or a host device.

Hardware based status collector acceleration engine for memory sub-system operations

A data bus coupled to a plurality of memory devices is determined to be in a read mode. Responsive to determining that the data bus is in the read mode, a particular read operation identified in a particular memory queue of memory queues that include identifiers of one or more write operations and identifiers of one or more read operations is determined. The particular memory queue includes a highest number of read operations for a memory device of the memory devices. The particular read operation is transmitted from the particular memory queue over the data bus.

Scheduling of read operations and write operations based on a data bus mode

A system includes a plurality of memory dice and a processing device coupled to the plurality of memory dice. The processing device is to determine whether an error correcting code (ECC) check of ECC-protected data read from a die of the plurality of memory dice results in detecting an error. In response to detecting the error from the ECC-protected data, the processing device performs a confirmation check that the error is a result of a defect in the die. In response to the confirmation check confirming the die is defective, the processing device ignores a temperature value from the die when determining whether to trigger a thermal-related operation.

Detect whether die or channel is defective to confirm temperature data

A system for managing power loss can include a number of memory devices including a volatile memory device and a non-volatile memory device and a processing device operatively coupled with the plurality of memory devices. The processor can save a snapshot of a logical-to-physical (L2P) table to a non-volatile memory device and maintain a journal of updates of the L2P. The processor can retrieve a sequence number from system metadata and save the most recent set of updates of the L2P table to a dedicated area of the non-volatile memory device, where the dedicated area is identified by the sequence number.

Managing power loss in a memory device

A method includes generating parity data corresponding to a plurality of word lines coupled to blocks of a memory device and generating additional parity data for a block based on a physical location of the block. The method can further include performing a data recovery operation based on the parity data, the additional parity data, or a combination thereof.

Proximity based parity data management

Systems and methods are disclosed including a memory device and a processing device operatively coupled to the memory device. The processing device can perform operations comprising performing a data integrity check on a set of memory cells of a source management unit of the memory device to obtain a data integrity metric value; determining whether the data integrity metric value satisfies a first threshold; responsive to determining that the data integrity metric value fails to satisfy the first threshold, determining whether the data integrity metric value satisfies a second threshold that is lower than the first threshold; responsive to determining that the data integrity metric value satisfies the second threshold, causing the memory device to copy data from the source management unit to a destination set of pages of the memory device; and performing a subsequent data integrity check on one or more invalid pages of the source management unit.

Performing block-level media management operations for block stripes in a memory device

A processing device in a memory system determines that a number of commands from an active queue that have been executed on a memory device does not satisfy an executed transaction threshold criterion, that a number of pending commands in an inactive queue satisfies a first promotion threshold criterion, and that a number of pending commands in the active queue does not satisfy a second promotion threshold criterion. In response, the processing device switches an execution grant from the active queue to the inactive queue.

Double threshold controlled scheduling of memory access commands

A set of blocks of a memory device comprising a plurality of dies is identified. A block within the set of blocks is identified. The identified block is associated with a capability metric that reflects a projected reliability of the die on which the block resides. Responsive to determining that the capability metric satisfies a condition, a cycle threshold associated with the die is identified. Responsive to determining that a cycle count value derived from a program/erase cycle counter associated with the die matches the cycle threshold, the set of blocks is updated by excluding the block from the set of blocks. A program operation is performed with respect to the updated set of blocks.

A current cycle count associated with a memory sub-system is determined. The current cycle count is compared to a set of cycle count threshold levels to determine a current lifecycle stage of the memory sub-system. A temperature associated with the memory sub-system is measured. The temperature is compared to a set of temperature levels to determine a current temperature level of the memory sub-system. A write-to-read delay time corresponding to the current lifecycle stage and the current temperature level is determined.

Managing an adjustable write-to-read delay based on cycle counts in a memory sub-system

A data bus is determined to be in a write mode. Whether a number of memory queues that identify at least one write operation satisfies a threshold criterion is determined. The memory queues include identifiers of one or more write operations and identifiers of one or more read operations. Responsive to determining that the number of memory queues satisfies the threshold criterion, a write operation from the memory queues is transmitted over the data bus.

A system for managing power loss can include a number of memory devices including a volatile memory device and a non-volatile memory device and a processing device operatively coupled with the plurality of memory devices. The processor can save a snapshot of a logical-to-physical (L2P) table to a non-volatile memory device and maintain a journal of updates of the L2P. The processor can, in response to a parameter of journal buffer of a volatile memory device satisfying a threshold criterion, save at least one journal of updates of the L2P table to the non-volatile memory device. It can also retrieve a sequence number from system metadata and save the most recent set of updates of the L2P table to a dedicated area of the non-volatile memory device, where the dedicated area is identified by the sequence number, in response to detecting a power loss event.

Methods, systems, and devices for memory can include techniques for identifying first quantities of write counts for a first plurality of super management units (SMUs) in a mapped region of a memory sub-system, identifying, by a hardware component of the memory sub-system, a first SMU of the first plurality that includes a fewest quantity of write counts of the first quantity of write counts, and performing a wear-leveling operation based at least in part on a first quantity of write counts of the first SMU of the first plurality in the mapped region being less than a second quantity of writes counts of a second SMU of a second plurality of SMUs in an unmapped region of the memory sub-system.

Growing community of inventors

Dublin, CA, United States of America

Wei Wang