A starting read threshold is received. A first offset and a second offset is determined. A first read is performed at the starting read threshold offset by the first offset to obtain a first hard read value and a second read is performed at the starting read threshold offset by the second offset to obtain a second hard read value. A soft read value is generated based at least in part on the first hard read value and the second hard read value.

This disclosure provides for improvements in managing multi-drive, multi-die or multi-plane NAND flash memory. In one embodiment, the host directly assigns physical addresses and performs logical-to-physical address translation in a manner that reduces or eliminates the need for a memory controller to handle these functions, and initiates functions such as wear leveling in a manner that avoids competition with host data accesses. A memory controller optionally educates the host on array composition, capabilities and addressing restrictions. Host software can therefore interleave write and read requests across dies in a manner unencumbered by memory controller address translation. For multi-plane designs, the host writes related data in a manner consistent with multi-plane device addressing limitations. The host is therefore able to “plan ahead” in a manner supporting host issuance of true multi-plane read commands.

Embodiments provide a scheme for parametric PV-level modeling and an optimal read threshold voltage estimation in a memory system. A controller performs read operations on cells using read threshold voltages; generates CMF samples based on the read operations; and receives first and second CDF values, which correspond to CMF samples, each CDF value representing a skew normal distribution. The controller estimates first and second probability distribution parameter sets corresponding to the first and second CDF values, respectively; determines first and second PDF values using the first and second probability distribution parameter sets, respectively; and estimates, as an optimal read threshold voltage, a read threshold voltage corresponding to a cross-point of the first and second PDF values.

Techniques described and suggested herein include systems and methods for optimizing performance characteristics by differentiating data storage device types for data archives stored on data storage systems using redundancy coding techniques. For example, redundancy coded shards, which may include identity shards that contain unencoded original data of archives, may be stored on different types of data storage devices to optimize for various retrieval use cases and implemented environments. Implementing systems may monitor various performance characteristics so as to adaptively account for changes to some or all of the monitored parameters.

Methods, systems, and devices for extended error detection for a memory device are described. For example, during a read operation, the memory device may perform an error detection operation capable of detecting single-bit errors, double-bit errors, and errors that impact more than two bits and indicate the detected error to a host device. The memory device may use parity information to perform an error detection procedure to detect and/or correct errors within data retrieved during the read operation. In some cases, the memory device may associate each bit of the data read during the read operation with two or more bits of parity information. For example, the memory device may use two or more sets of parity bits to detect errors within a matrix of the data. Each set of parity bits may correspond to a dimension of the matrix of data.

Techniques for encoding data for non-volatile memory storage systems are disclosed. In one particular embodiment, the techniques may be realized as a method including writing first data to the memory, reading the first data from the memory, analyzing the first read data such that the analyzing includes determining whether the read data includes an error, encoding second data based on the analyzing of the first data such that the second data is encoded to be written to a position adjacent to the error when it is determined that the read data includes the error, and writing the encoded second data to the memory at the position.

A decoding method, a memory storage device and a memory control circuit unit are provided. The decoding method includes: reading memory cells based on a default hard-decision voltage level and decoding the obtained hard-bit information; if the decoding fails, reading the memory cells based on default soft-decision voltage levels and then decoding the obtained soft-bit information; if the decoding still fails, reading the memory cells based on first test voltage levels to obtain first soft-bit information and reading the memory cells based on second test voltage levels to obtain second soft-bit information; obtaining a first estimating parameter and a second estimating parameter according to the first soft-bit information and the second soft-bit information, respectively; and updating the default hard-decision voltage level according to the first estimating parameter and the second estimating parameter. As a result, a decoding efficiency can be improved.

A semiconductor memory device includes a memory cell, a word line connected to the memory cell, a source line connected to the memory cell, a bit line connected to the memory cell, and a control circuit configured to perform a read operation on the memory cell. During the read operation, the control circuit applies to the word line a first voltage, a second voltage greater than the first voltage after applying the first voltage, and a third voltage greater than the first voltage and smaller than the second voltage after applying the second voltage, and applies to the source line a fourth voltage according to a timing at which the second voltage is applied to the word line, a fifth voltage smaller than the fourth voltage after applying the fourth voltage, and a sixth voltage greater than the fifth voltage after applying the fifth voltage.

According to one embodiment, in a memory system, a controller is configured to write first data in a page in a block in response to a write request from a host, and update second information used to manage a correspondence between a logical address designated by the write request and a second physical address which is a storage location in the first memory. The controller is configured to perform a first process of updating the first information with the second information and storing the updated information in the first memory. The controller is configured to acquire the first physical address associated to a logical address designated by the write request from the first information. The controller is configured to store, in the first memory, third information including information in which the acquired first physical address and the second physical address are associated.

A flash memory storage management method includes: providing a flash memory module including single-level-cell (SLC) blocks and at least one multiple-level-cell block such as MLC block, TLC block, or QLC block; classifying data to be programmed into groups of data; respectively executing SLC programming and RAID-like error code encoding to generate corresponding parity check codes, to program the groups of data and corresponding parity check codes to the SLC blocks; when completing program of the SLC blocks, performing an internal copy to program the at least one multiple-level-cell block by sequentially reading and writing the groups of data and corresponding parity check codes from the SLC blocks to the multiple-level-cell block according to a storage order of the SLC blocks.