摘要:
Apparatuses, systems, methods, and computer program products are disclosed for storage operations for a non-volatile medium. A control module may be configured to divide a storage procedure into multiple portions. An execution module may be configured to execute multiple portions of a storage procedure independently. A storage request module may be configured to satisfy a storage request for one or more storage elements of a storage procedure between at least a pair of portions of a storage procedure.
摘要:
A storage module is configured to store data segments, such as error-correcting code (ECC) codewords, within an array comprising two or more solid-state storage elements. The data segments may be arranged in a horizontal arrangement, a vertical arrangement, a hybrid channel arrangement, and/or vertical stripe arrangement within the array. The data arrangement may determine input/output performance characteristics. An optimal adaptive data storage configuration may be based on read and/or write patterns of storage clients, read time, stream time, and so on. Data of failed storage elements may be reconstructed by use of parity data and/or other ECC codewords stored within the array.
摘要:
Apparatuses, systems, methods, and computer program products are provided for error correction. A soft read module is configured to obtain soft read information for a cell of a non-volatile memory medium. The soft read information may indicate a likelihood that a data value for the cell is correct. A reliability module is configured to associate the cell with a log-likelihood ratio (LLR) mapping from a plurality of LLR mappings based on one or more reliability characteristics for a set of cells that includes the cell. An LLR map module is configured to determine an LLR value based on the soft read information by using the LLR mapping.
摘要:
Apparatuses, systems, methods, and computer program products are disclosed for error correcting code (ECC) decoding. A score module is configured to assign a score to a variable node of low density parity check (LDPC) code decoder. The LDPC code decoder may include a plurality of variable nodes associated with a plurality of check nodes. The plurality of variable nodes may correspond to bits of a received code word. The score for the variable node may be based on a count of unsatisfied check nodes associated with the variable node. A variable node update module is configured to update the variable node based on the score. A check node update module is configured to update one or more of the check nodes based on the updated variable node.
摘要:
A storage controller may be configured to assess the reliability of a solid-state storage medium. The storage controller may be further configured to project, forecast, and/or estimate storage reliability at a future time. The projection may be based on a currently reliability metric of the storage and a reliability model. The portions or sections of the solid-state storage media may be retired in response the projected reliability metric failing to satisfy a reliability threshold. The reliability threshold may be based on data correction and/or reconstruction characteristics. The projected reliability metrics of a plurality of erase blocks of a storage division may be combined, and one or more of the erase blocks may be retired in response to determining that the combined reliability metric projection fails to satisfy the reliability threshold.
摘要:
Apparatuses, systems, methods, and computer program products are disclosed for error correction. A soft read module is configured to obtain soft read information for a cell of a non-volatile memory medium. The soft read information may indicate a likelihood that a data value for the cell is correct. A reliability module is configured to associate the cell with a log-likelihood ratio (LLR) mapping from a plurality of LLR mappings based on one or more reliability characteristics for a set of cells that includes the cell. An LLR map module is configured to determine an LLR value based on the soft read information by using the LLR mapping.
摘要:
Apparatuses, systems, and methods are disclosed for a virtual channel for data transfers between devices. A method includes presenting an address space for a memory buffer. The address space may be larger than a physical capacity of the memory buffer. A method includes controlling, from a peripheral device, a rate at which a data source transfers data to a memory buffer using a presented address space based on a rate at which a data target transfers data from the memory buffer using the presented address space so that an amount of data stored in the memory buffer remains at or below a physical capacity of the memory buffer.
摘要:
A storage module is configured to determine a health metric of a storage division of a solid-state storage medium. The health metric may comprise a combination of factors, including, but not limited to: wear level, performance (e.g., program time, erase time, and the like), error rate, and the like. A wear level module may configure storage operations to reduce the wear rate of storage divisions having poor health metrics and/or heath metrics that are degrading more quickly than other storage divisions. Reducing wear rate may include deferring grooming operations, delaying use for storage operations, temporarily retiring the storage division, or the like. Storage divisions may be brought back into service at normal use rates in response determining that other portions of the storage media have been worn to the point that they exhibit similar health and/or reliability characteristics.
摘要:
A memory device includes a memory array with a plurality of memory elements. Each memory element is configured to store data. The device includes an input/output (I/O) buffer coupled to the memory array. The I/O buffer is configured to receive data from an I/O interface of a memory device controller and write the data to the memory array. The device includes a memory control manager coupled to the memory array. The memory control manager is configured to pause a program operation to the memory array in response to receiving a pause command. The memory control manager is also configured to resume the program operation in response to receiving a resume command.
摘要:
Apparatuses, systems, and methods are disclosed for managing non-volatile a medium. A method includes determining whether a first error correcting code (ECC) code word of a non-volatile storage device is correctable using a first error correcting code. A method includes determining whether a second ECC code word is correctable using a second error correcting code in response to determining that a first ECC code word is uncorrectable using a first error correcting code. A method includes adjusting one or more media parameters for accessing a non-volatile medium of a non-volatile storage device based on error information. Error information may include information from a decoder for a second error correcting code. Adjusting one or more media parameters may be in response to determining that a second ECC code word is correctable using a second error correcting code.