Abstract: Data that is stored in a higher error rate format in a nonvolatile memory is backed up in a lower error rate format. Data to be stored may be transferred once to on-chip data latches where it is maintained while it is programmed in both the high error rate format and the low error rate format without being resent to the nonvolatile memory. High error rate format may be MLC format and programming in the high error rate format may program both lower page and upper page data together in a full sequence programming scheme that is suitable for handling high data volume.

Abstract: Data that is stored in a higher error rate format in a nonvolatile memory is backed up in a lower error rate format. Data to be stored may be transferred once to on-chip data latches where it is maintained while it is programmed in both the high error rate format and the low error rate format without being resent to the nonvolatile memory.

Abstract: In a nonvolatile memory array in which a NAND string includes a back gate that has a charge storage element, the threshold voltage of the back gate is monitored, and if the threshold voltage deviates from a desired threshold voltage range, charge is added to, or removed from the charge storage element to return the threshold voltage to the desired threshold voltage range.

Abstract: Data that is stored in a higher error rate format in a 3-D nonvolatile memory is backed up in a lower error rate format. Later, the higher error rate copy is sampled to determine if it is acceptable. A sampling pattern samples all word lines of a string and at least one word line of each string of the block.

Abstract: When data from a portion of a three dimensional NAND memory array is determined to be uncorrectable by Error Correction Code (ECC), a determination is made as to whether data is uncorrectable by ECC throughout some unit that is larger than the portion. If modified read conditions provide ECC correctable data, the modified read conditions are recorded for subsequent reads of the larger unit.

Abstract: Portions of data stored in a three dimensional memory array are selected based on their locations for calculation of redundancy data. Locations are selected so that no two portions in a set of portions for a given calculation are likely to become uncorrectable at the same time. Selected portions may be separated by at least one word line and separated by at least one string in a block.

Abstract: In a multi-level cell memory array, a flag that indicates that a logical page is unwritten is subject to a two-step verification. In a first verification step, the logical page is read, and ECC decoding is applied. If the first verification step indicates that the logical page is unwritten, then a second verification step counts the number of cells that are not in an unwritten condition.

Abstract: In a multi-level cell (MLC) nonvolatile memory array, data is assigned sequentially to the lower and upper page of a word line, then both lower and upper pages are programmed together before programming a subsequent word line. Word lines of multiple planes are programmed together using latches to hold data until all data is transferred. Tail-ends of data of write commands are stored separately.

Abstract: In a non-volatile memory device, the parameters used in write and erase operation are varied based upon device age. For example, in a programming operation using a staircase waveform, the amplitude of the initial pulse can be adjusted based upon the number of erase-program cycles (hot count) of the block containing the selected physical page for the write. This arrangement can preserve performance for relatively fresh devices, while extending life as a devices ages by using gentler waveforms as the device ages.

Abstract: A non-volatile memory system receives and stores host data. As the memory system receives host data, it initially writes the data in a binary format and then subsequently performs an on-chip folding operation on the data, storing the data in a multi-state format. The memory system interleaves the phases of the folding operations, where the host stores data according to allocation units. The memory system also can perform the binary and subsequent on-chip folding operations on multiple memory planes in parallel, where the controller also balances the operations. When the memory system needs a memory block for the writing of data, it selects blocks from a free block list, where the free block list includes a reserve portion that is only accessible for a specified set of commands.

Abstract: Memory cells that are indicated as being erased but are suspected of being partially programmed may be subject to a verification scheme that first performs a conventional read and then, if the conventional read does not indicate partial programming, performs a second read using lower read-pass voltage on at least one neighboring word line.

Abstract: Lower page data that may be endangered by programming upper page data in the same memory cells is protected during upper programming using protective upper page programming schemes. High overall programming speeds are maintained by selectively using protective upper programming schemes only where endangered data is committed and may not be recoverable from another location.

Abstract: In a multi-level cell memory array, a flag that indicates that a logical page is unwritten is subject to a two-step verification. In a first verification step, the logical page is read, and ECC decoding is applied. If the first verification step indicates that the logical page is unwritten, then a second verification step counts the number of cells that are not in an unwritten condition.

Abstract: Data errors in non-volatile memory inevitably increase with usage and with higher density of bits stored per cell. The memory is configured to have a first portion operating with less error but of lower density storage, and a second portion operating with a higher density but less robust storage. Input data is written and staged in the first portion before being copied to the second portion. An error management provides checking the quality of the copied data for excessive error bits. The copying and checking are repeated on a different location in the second portion until either a predetermined quality is satisfied or the number or repeats exceeds a predetermined limit. The error management is not started when a memory is new with little or no errors, but started after the memory has aged to a predetermined amount as determined by the number of erase/program cycling its has experienced.

Abstract: A non-volatile memory system receives and stores host data. As the memory system receives host data, it initially writes the data in a binary format and then subsequently performs an on-chip folding operation on the data, storing the data in a multi-state format. The memory system interleaves the phases of the folding operations so that performance is made more uniform across allocation units, where the host stores data according to allocation units. The memory system also can perform the binary and subsequent on-chip folding operations on multiple memory planes in parallel, where the controller also balances the operations so that performance is made more uniform between planes with respect to allocation units as the data is received from the host. To further maintain performance, the memory system uses a free block list having a reserve portion that is only accessible for a specified set of commands.

Abstract: A program operation in a non-volatile memory is segmented at predefined junctures into smaller segments for execution over different times. The predefined junctures are such that they allow unambiguous identification when restarting the operation in a next segment so that the operation can continue without having to restart from the very beginning of the operation. This is accomplished by requiring the programming sequence of each segment to be atomic, that is, to only terminate at a predetermined type of programming step. In a next segment, the terminating programming step is identified by detecting a predetermined pattern of ECC errors across a group of programmed wordlines.

Abstract: A program operation in a non-volatile memory is segmented at predefined junctures into smaller segments for execution over different times. The predefined junctures are such that they allow unambiguous identification when restarting the operation in a next segment so that the operation can continue without having to restart from the very beginning of the operation. This is accomplished by requiring the programming sequence of each segment to be atomic, that is, to only terminate at a predetermined type of programming step. In a next segment, the terminating programming step is identified by detecting a predetermined pattern of ECC errors across a group of programmed wordlines.

Abstract: Data errors in non-volatile memory inevitably increase with usage and with higher density of bits stored per cell. The memory is configured to have a first portion operating with less error but of lower density storage, and a second portion operating with a higher density but less robust storage. Input data is written and staged in the first portion before being copied to the second portion. An error management provides checking the quality of the copied data for excessive error bits. The copying and checking are repeated on a different location in the second portion until either a predetermined quality is satisfied or the number or repeats exceeds a predetermined limit. The error management is not started when a memory is new with little or no errors, but started after the memory has aged to a predetermined amount as determined by the number of erase/program cycling its has experienced.