A memory controller includes a fault predictor which predicts a fault which causes an error occurring in a memory device, an error correction code (ECC) manager which classifies a type of the fault based on the predicted fault, and a plurality of ECC engines which perform ECC in parallel depending on the classified type of the faults. The fault predictor includes a memory error profiler which receives raw data related to the error and processes the raw data into an error profile that is data available for machine learning, and a memory fault prediction network which receives the error profile as an input, performs the machine learning using the error profile, and predicts the fault which causes the error.

Performing refresh operation in a memory device is provided. A refresh operation without address rotation is performed in a cell array of the memory device. Performing the refresh operation without address rotation is repeated for a predetermined number of times. After repeating performing the refresh operation with address rotation for the predetermined number of times, a refresh operation with address rotation is performed in the cell array.

A solid state drive (SSD) employing a redundant array of independent disks (RAID) scheme includes a flash memory chip, erasable blocks in the flash memory chip, and a flash controller. The erasable blocks are configured to store flash memory pages. The flash controller is operably coupled to the flash memory chip. The flash controller is also configured to organize certain of the flash memory pages into a RAID line group and to write RAID line group membership information to each of the flash memory pages in the RAID line group.

Examples relate to a concept for handling transient errors. An apparatus for correcting transient errors in a computational device comprises interface circuitry, machine-readable instructions and processing circuitry for executing the machine-readable instructions to obtain a signal indicating that a transient error has been detected in the computational device, the computational device being configured to perform computations using processing elements and connections between the processing elements, extract a state of the computational device, the state comprising at least one of present and previous values transmitted via the connections between the processing elements and state contained within the one or more processing elements, compute a corrected state of the computational device based on the state extracted from the computational device, and configure a computational device with the corrected state.

A storage system determines that it is undergoing intensive reads by a host, which can occur, for example, when the storage system is being used to play a video game for a prolonged period of time. As performing a conventional read scrub operation in that situation can result in a decrease in performance, the storage system can instead use a targeted read scrub operation to reduce the impact on host read performance. The targeted read scrub operation can take the form, for example, of a periodic read scan on areas of the memory that are not part of the intensive host read, random read scans on neighboring wordlines where only a single state is read, and/or a passive read scan where acceptable but risky pages are marked for relocation.

Embodiments of the present disclosure relate to a memory system and an operating method thereof. According to the embodiments of the present disclosure, the memory system may completely scan each of one or more target memory blocks among the plurality of memory blocks, once in each scan period to detect an error in data stored in the corresponding target memory block and may block an attempted second scan of each target memory block in a scan period in which the corresponding target memory block has already been scanned until the scan period is completed.

Methods, devices, and systems related to memory device on-die ECC data are described. In an example, a scrub operation can be performed on data in order to determine which rows of memory cells in an array include a particular number of errors. The particular number of errors can be a number of errors that exceed a threshold number of errors. An address of the determined rows with the particular number of errors can be stored in memory cells of the array for later access. The address of the determined rows can be accessed to perform a user-initiated repair operation, a self-repair operation, a refresh operation, and/or to alter timing of access of the cells or alter voltage of the cells.

A semiconductor system includes a controller configured to count the number of error check scrub (ECS) operations and configured to generate ECS information that includes information with regard to an address at which the ECS operation is to be performed based on the number of ECS operations. The semiconductor system further includes a memory apparatus configured to perform the ECS operation on a region that is selected by the ECS information.

A memory device including: a memory cell array including a plurality of memory cells disposed at intersections of wordlines and bitlines; an error correction circuit configured to read data from the memory cell array and to correct an error in the read data; and an error check and scrub (ECS) circuit configured to perform a scrubbing operation on the memory cell array, wherein the ECS circuit includes: a first register configured to store an error address obtained in the scrubbing operation; and a second register configured to store a page offline address received from an external device.

A semiconductor memory device includes a memory cell array, an error correction code (ECC) circuit, a fault address register, a scrubbing control circuit and a control logic circuit. The memory cell array includes memory cell rows. The scrubbing control circuit generates scrubbing addresses based on refresh operations performed on the memory cell array. The control logic circuit controls the ECC circuit such that the ECC circuit performs an error detection operation on a plurality of sub-pages in a first memory cell row to count a number of error occurrences, and determines whether to correct a codeword in which an error is detected based on the number of error occurrences. An uncorrected or corrected codeword is written back, and a row address of the first memory cell row may be stored in the fault address register as a row fault address based on the number of error occurrences.