Micron Technology Incorporated

Via Technologies, Inc.

Sandisk Enterprise IP LLC

Sandisk Technologies Inc.

Jiangli Zhu

Ying Yu Tai

Fangfang Zhu

Wei Wang

Tingjun Xie

Ning Chen

Chih-Kuo Kao

Juane Li

Zhenming Zhou

Seungjune Jeon

Murong Lang

Jason Duong

Jian Huang

Yueh-Hung Chen

Zhengang Chen

Cheng Yuan Wu

Kishore Kumar Muchherla

Sanjay Subbarao

Akira Goda

Samir Mittal

Xiaoheng Chen

Dave Scott Ebsen

Frederick Adi

Vivek Shivhare

Venkata Naga Lakshman Pasala

Peter Sean Feeley

Zhenlei Shen

Alex G Tang

Lakshmi Kalpana Vakati

Zhongguang Xu

Huapeng G Guan

Yipei Yu

Jiin Lai

Horia Cristian Simionescu

Yang Liu

Yang Lu

Nagendra Prasad Ganesh Rao

Aaron Lee

Yi-Min Lin

Paul Roger Stonelake

Charles S Kwong

Sead Zildzic, Jr

Sead Zildzic

Jingyu Kang

Ying Yu Kai

Nagireddy Chodem

Yuehhung Chen

Ying-Yu Tai

Methods, systems, and apparatuses include receiving a write command including user data. The write command is directed to a portion of memory including a first block and a second block. A buffer is allocated for executing the write command to the first block. The buffer includes multiple buffer decks and the buffer holds the user data written to the first block. User data is programmed into the first block to a threshold percentage. The threshold percentage is less than one hundred percent of the first block. A buffer deck is invalidated in response to programming the first block to the threshold percentage. The buffer deck is reallocated to the second block for programming the user data into the second block. The buffer deck holds user data written to the second block.

Buffer allocation for reducing block transit penalty

A processing device in a memory sub-system traverses a plurality of management units of a memory device at a defined scan/read refresh frequency. For every management unit of the plurality of management units, the processing device identifies a page satisfying a lowest sensing overhead criterion, and senses data of the identified page without transferring the data out of the memory device. A non-transitory computer readable medium includes program instructions that when executed by a processing device, cause the processing device to perform operations of traversing a plurality of management units of a memory device at a defined scan/read refresh frequency. For every management unit, the processing device identifies a page satisfying a lowest sensing overhead criterion, and senses data of the identified page without transferring the data out of the memory device.

Efficient periodic backend refresh reads for reducing bit error rate in memory devices

Methods, apparatuses and systems related to managing read levels across multiple groupings or decks are described. The apparatus may include a read level management mechanism configured to controls or dynamically adjusts the read levels for the different groupings according to a real-time condition/measure associated with the programming delays.

Apparatus with multi-deck read level management and methods for operating the same

Methods, systems, and apparatuses include receiving a write command including user data. The write command is directed to a portion of memory including a first and second block and a first and second user data portion are directed to the first and second block. Temporary parity data is generated using the first and second user data portions. The temporary parity data and the first and second user data portions are stored in a buffer. Portions of the first and second block are programmed with two programming passes. The first and second user data portions in the buffer are invalidated in response to a completion of the second programming pass of the portions of the first and second blocks. The temporary parity data is maintained in the buffer until a second programming pass of the first and second block.

Multiple-pass programming of memory cells using temporary parity generation

A memory sub-system periodically performs a first wear leveling operation using a direct mapping function on a data management unit of a memory component in the memory sub-system at a first frequency. The memory sub-system further periodically performs a second wear leveling operation using indirect mapping on a group of data management units of the memory component at a second frequency, wherein the second wear leveling operation is performed less frequently than the first wear leveling operation.

Hybrid wear leveling for in-place data replacement media

A determination is made of whether a memory sub-system operates in a full capacity mode or a reduced capacity mode. The full capacity mode corresponds to accessing data residing at a set of memory devices via a number of physical data channels that corresponds to a number of logical data channels. The reduced capacity mode corresponds to accessing the data via a number of physical data channels that is less than the number of logical data channels. A data structure is updated to include one or more mappings between physical data channels and logical data channels according to the determination. A memory access operation to access a data item at memory cells of at least one of the set of memory devices is executed based on the one or more mappings of the data structure.

Enabling multiple data capacity modes at a memory sub-system

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 threshold criterion of a plurality of threshold criteria is identified based on a current program-erase cycle (PEC) count of a first block of a memory device, wherein the first block is configured as quad-level cell (QLC) memory. A raw bit error rate (RBER) associated with data of a second block of the memory device is determined, wherein the second block is configured as single-level cell (SLC) memory. It is determined that the RBER associated with the data of the second block satisfies the threshold criterion. In response to determining that the RBER satisfies the threshold criterion, the data of the second block is written to the first block.

Managing an adaptive data path selection threshold for a memory sub-system

A stripe-based command pertaining to a set of host data items at management units (MUs) of a memory sub-system configured to support non-stripe based commands is received. A set of operations to be executed at the MUs based on the stripe-based command is determined. The set of operations include one or more first operations associated with the set of host data items, the one or more first operations having a first type, and one or more second operations associated with the set of host data items, the one or more second operations having a second type. A first set of commands corresponding to the one or more first operations and a second set of commands corresponding to the one or more second operations is executed.

Enabling stripe-based operations for error recovery at a memory sub-system

A processing device in a memory sub-system performs a first media scan operation with respect to a plurality of memory pages addressable by the ordinary wordline, wherein each page of the plurality of memory pages is contained by a respective management unit, and responsive to determining that a value of a data state metric of a memory page of the plurality of memory page addressable by the ordinary wordline satisfies a specified condition, performs a first media management operation with respect to a management unit containing the memory page. The processing device further performs a second media scan operation with respect to a plurality of memory pages addressable by the mandatory wordline, wherein each page of the plurality of memory pages is contained by the respective management unit, and responsive to determining that a value of the data state metric of a memory page of the plurality of memory page addressable by the mandatory wordline satisfies the specified condition, performs a second media management operation with respect to the management unit containing the memory page.

Reducing bit error rate in memory devices

A partition command is stored at free memory address location of the local memory corresponding to an index of an address array. The index is associated with an entry in the address array. A last entry in a linked list of entries from a tail register is obtained based on an allocation of the stored partition command to a partition command queue of a plurality of partition command queues. The tail register corresponds to the partition command queue of the plurality of partition command queues. Responsive to obtaining the last entry in the linked list, an entry to the linked list after the last entry is appended. The entry corresponds to the index of the address array associated with the stored partition command.

Dynamic partition command queues for a memory device

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

Disclosed is a system comprising a memory device and a processing device, operatively coupled with the memory device, to perform operations including identifying a group of memory cells corresponding to a first range of logical block addresses (LBAs). The operations performed by the processing device further include receiving a memory access command with respect to the group of memory cells. The operations performed by the processing device further include responsive to determining that a data structure associated with the group of memory cells references a second range of LBAs, blocking the memory access command; responsive to determining that the first range of LBAs does not include each LBA of the second range of LBAs, performing, on the group of memory cells, a trim operation; and responsive to determining that the data structure indicates the completion of the trim operation, performing a memory access operation specified by the memory access command.

Managing trim commands in a memory sub-system

A difference between a recorded time stamp for a first set of memory cells comprised by an open translation unit (TU) of memory cells and a current time stamp for the open TU is determined, wherein the first set of memory cells comprises a most recently programmed set of memory cells. It is determined, based on a current temperature for the open TU and the difference between the recorded time stamp and the current time stamp, that a second set of memory cells comprised by the open TU is in a coarse programming state. A programming operation is performed on the second set of memory cells using a reduced programming state verify level and a reduced programming state gate step size associated with the second set of memory cells.

Managing the programming of an open translation unit

Embodiments include methods, systems, devices, instructions, and media for limiting hot-cold swap wear leveling in memory devices. In one embodiment, wear metric values are stored and monitored using multiple wear leveling criteria. The multiple wear leveling criteria include a hot-cold swap wear leveling criteria, which may make use of a write count offset value. Based on a first wear metric value of a first management group and a second wear metric value of a second management group, the first management group and the second management group are selected for a wear leveling swap operation. The wear leveling swap operation is performed with a whole management group read operation of the first management group to read a set of data, and a whole management group write operation to write the set of data to the second management group.

Limiting hot-cold swap wear leveling

Growing community of inventors

San Jose, CA, United States of America

Jiangli Zhu