Abstract: A method including mapping an address space of the buffer configured to store a plurality of data values into a first two-dimensional array of values. For each row in the first two-dimensional array, calculating, by a processor comprising an encoder, a row parity value. For each row, a plurality of data values in the row and the row parity value form a row codeword. For each column in the first two-dimensional array, a column parity value is calculated by the processor comprising an encoder, wherein for each column, a plurality of data values in the column and the column parity value form a column codeword. The exclusive-OR (XOR) of the plurality of data values is calculated. A parity value based on the XOR of the plurality of data values is calculated by the processor comprising an encoder.

Abstract: Log likelihood ratio (LLR) values that are computed in a flash memory controller during read retries change over time as the number of program-and-erase cycles (PECs) that the flash memory die has been subjected to increases. Therefore, in cases where an LLR table is used to provide pre-defined, fixed LLR values to the error-correcting code (ECC) decoding logic of the controller, decoding success and the resulting BER will degrade over time as the number of PECs to which the die has been subjected increases. In accordance with embodiments, a storage system, a flash memory controller for use in the storage system and method are provided that periodically measure the LLR values and update the LLR table with new LLR values. Periodically measuring the LLR values and updating the LLR table with new LLR values ensures high decoding success and a low BER over the life of the flash memory die.

Abstract: An apparatus includes an interface and a control circuit. The interface may be configured to process read/write operations to/from a memory. The control circuit may be configured to create dependencies between a current bit in a sequence of data bits and neighboring bits in the sequence of data bits to generate mapped bits in response to a condition in a region of the memory being true, write the mapped bits among at least two memory cells in the region of the memory with at least two of the mapped bits stored in each of the memory cells, where the dependencies mitigate a hard error due to one of the at least two cells being stuck in a fixed state, and write the sequence of data bits in the region of the memory in response to the condition in the region of the memory being false.

Abstract: An apparatus includes a memory and a controller. The memory may be configured to store data. The memory generally comprises a plurality of memory units each having a size less than a total size of the memory. The controller may be configured to generate a set of converted log likelihood ratios by scaling a set of original log likelihood ratios using a selected scalar value, wherein the controller determines the selected scalar value by generating a plurality of sets of scaled log likelihood ratios by scaling the set of original log likelihood ratios with a plurality of corresponding scalar values, calculating a plurality of respective correlation coefficients each measuring a similarity of a respective set of scaled log likelihood ratios to the set of original log likelihood ratios, and selecting the scalar value corresponding to the set of scaled log likelihood ratios whose respective correlation coefficient is highest as the selected scalar value.

Abstract: Systems and method relating generally to solid state memory, and more particularly to systems and methods for generated data from a solid state memory. A data processing system includes a solid state memory device, a soft data generation circuit operable to receive multiple instances of an element of a read data set accessed from the solid state memory device, and access a scramble compensating extended look up table using the multiple instances of the element to receive corresponding soft data, and a data decoder circuit operable to yield a decoded output from the soft data.

Abstract: An apparatus comprises a hybrid-memory multi-chip package (MCP) including a non-volatile memory (NVM) in an NVM die; a dynamic random access memory (DRAM) in two or more DRAM die, wherein a portion of the DRAM is allocated as a cache memory for the NVM; and a hybrid controller for the NVM and DRAM. The hybrid controller includes an NVM interface to the NVM; a DRAM interface to the cache memory; a host interface to communicate data with a host processor, wherein the host interface includes a parallel data bus for reading and writing data directly with both of the DRAM and the NVM; and logic circuitry configured to interleave access by the host processor and hybrid controller to the DRAM and NVM.

Abstract: A storage device implements a method for garbage collection. The storage device arranges data blocks of a storage medium into a bin and determines first coldness of a first data block in the bin and second coldness of a second data block in the bin that are respectively associated with a first rate of change of valid data in the first data block into invalid data and a second rate of change of valid data in the second data block into invalid data. Based on the first coldness and the second coldness, the storage device selects a colder data block from the first and second data blocks as a garbage data block. Because the valid data in the selected garbage data block are more stable, they may cause less new stale data or garbage data in a new block to which the valid data are moved.

Abstract: An apparatus comprising a memory and a controller. The memory may be configured to process a plurality of read/program operations. The memory may comprise a plurality of memory units. The memory units may each have a size less than a total size of the memory. The memory units may include a plurality of cells. The controller may be configured to issue a plurality of program operations to write to one or more of the cells. The controller may be configured to implement a polling status command after each of the program operations to verify programming of each of the cells. A response to each of the polling status commands may be used to report a number of the cells that failed to be programmed.

Abstract: Log likelihood ration (LLR) values that are computed in a flash memory controller during read retries change over time as the number of program-and-erase cycles (PECs) that the flash memory die has been subjected to increases. Therefore, in cases where an LLR table is used to provide pre-defined, fixed LLR values to the error-correcting code (ECC) decoding logic of the controller, decoding success and the resulting BER will degrade over time as the number of PECs to which the die has been subjected increases. In accordance with embodiments, a storage system, a flash memory controller for use in the storage system and method are provided that periodically measure the LLR values and update the LLR table with new LLR values. Periodically measuring the LLR values and updating the LLR table with new LLR values ensures high decoding success and a low BER over the life of the flash memory die.

Abstract: A method performed by a processor to improve wear-leveling in a cross-point (X3D) memory, comprises detecting, by a processor coupled to the X3D memory, a trigger event, wherein the X3D memory comprises a first section of memory units and a second section of memory units, and in response to detecting the trigger event, relocating, by the processor, data stored in a first memory unit of the first section of memory units to a memory unit adjacent to a last memory unit of the first section of memory units, and relocating, by the processor, data stored in a first memory unit of the second section of memory units to a memory unit adjacent to a last memory unit of the second section of memory units.

Abstract: A method is performed by a solid state device (SSD) controller to generate a parity. The method includes receiving input data to be stored to pages of a storage device, wherein each page is capable of being allocated with multiple codewords; configuring codewords of the pages into multiple groups, wherein each group has an integer number of codewords, at least one of the pages is allocated with a non-integer number of codewords, and wherein the integer number is larger than the non-integer number; obtaining parities for the multiple groups, and storing the parities to reserved spaces of the storage device. With the calculation of a parity decoupled from physical pages, the selection of a rate of code used is unconstrained by an integer number of codewords per page.

Abstract: An apparatus comprises a memory and a controller. The memory generally comprises a plurality of memory modules, each having a size less than a total size of the memory and configured to store data. The controller may be configured to process a plurality of read/write operations, classify data pages from multiple blocks of the memory as hot-read data or non hot-read data, and aggregate the hot-read data by selecting one or more of the hot-read data pages from multiple memory blocks and mapping the selected hot-read data pages to dedicated hot-read data blocks using a strong type of error correcting code during one or more of a garbage collection state, a data recycling state, or an idle state. The aggregation of the hot-read data pages and use of the strong type of error correcting code reduces read latency of the hot-read data pages, reduces a frequency of data recycling of the hot-read data pages, and reduces an impact of read disturbs on endurance of the memory.

Abstract: An apparatus comprises a hybrid-memory multi-chip package (MCP) including a non-volatile memory (NVM) in an NVM die; a dynamic random access memory (DRAM) in two or more DRAM die, wherein a portion of the DRAM is allocated as a cache memory for the NVM; and a hybrid controller for the NVM and DRAM. The hybrid controller includes an NVM interface to the NVM; a DRAM interface to the cache memory; a host interface to communicate data with a host processor, wherein the host interface includes a parallel data bus for reading and writing data directly with both of the DRAM and the NVM; and logic circuitry configured to interleave access by the host processor and hybrid controller to the DRAM and NVM.

Abstract: A method includes obtaining an average number of valid pages per block of the solid state storage device, obtaining an average number of invalid pages per block of the solid state storage device, determining a scaling factor as a function of the a number of free blocks in the solid state storage device, a steady state target number of free blocks, a high target number of free blocks, and a low target number of free blocks, and applying the scaling factor to the average number of invalid pages to control a ratio of host writes versus garbage collection writes to the solid state storage device.

Abstract: Systems and method relating generally to solid state memory, and more particularly to systems and methods for recycling data in a solid state memory. The systems and methods include receiving a data set maintained in a memory device, applying at least one iteration of a data decoding algorithm to the data set by a data decoder circuit to yield a decoded output, counting the number of iterations of the data decoding algorithm applied to the data set to yield an iteration count, and recycling the data set to the memory device. The recycling is triggered based at least in part on the iteration count.

Abstract: An apparatus comprising a memory and a controller. The memory is configured to process a plurality of read/write operations. The memory comprises a plurality of memory modules. Each memory module has a size less than a total size of the memory. The controller is configured to (i) classify data from multiple blocks of the memory as hot-read data or non hot-read data, (ii) aggregate the hot-read data to dedicated blocks, and (iii) select a type of error correcting code to protect the hot-read data in the dedicated blocks. The aggregation reduces an impact on endurance of the memory.

Abstract: Systems and methods relating generally to data processing, and more particularly to systems and methods for characterizing a solid state memory. In one embodiment, the systems and methods may include programming a first cell of a solid state memory device to a negative voltage, programming a second cell of the solid state memory device to a positive voltage, wherein the second cell is adjacent to the first cell, calculating a voltage shift on the negative voltage programmed to the first cell, characterizing a shifted voltage level on the first cell as an interim voltage, and subtracting the voltage shift from the interim voltage to yield an actual voltage on the first cell.

Abstract: Systems and method relating generally to solid state memory, and more particularly to systems and methods for recycling data in a solid state memory. The systems and methods include receiving a data set maintained in a memory device, applying at least one iteration of a data decoding algorithm to the data set by a data decoder circuit to yield a decoded output, counting the number of iterations of the data decoding algorithm applied to the data set to yield an iteration count, and recycling the data set to the memory device. The recycling is triggered based at least in part on the iteration count.

Abstract: Systems and methods relating generally to data processing, and more particularly to systems and methods for characterizing a solid state memory. In one embodiment, the systems and methods may include programming a first cell of a solid state memory device to a negative voltage, programming a second cell of the solid state memory device to a positive voltage, wherein the second cell is adjacent to the first cell, calculating a voltage shift on the negative voltage programmed to the first cell, characterizing a shifted voltage level on the first cell as an interim voltage, and subtracting the voltage shift from the interim voltage to yield an actual voltage on the first cell.

Abstract: An apparatus includes a memory and a controller. The memory may be configured to store data. The memory generally comprises a plurality of memory units each having a size less than a total size of the memory. The controller may be configured to generate a set of converted log likelihood ratios by scaling a set of original log likelihood ratios using a selected scalar value, wherein the controller determines the selected scalar value by generating a plurality of sets of scaled log likelihood ratios by scaling the set of original log likelihood ratios with a plurality of corresponding scalar values, calculating a plurality of respective correlation coefficients each measuring a similarity of a respective set of scaled log likelihood ratios to the set of original log likelihood ratios, and selecting the scalar value corresponding to the set of scaled log likelihood ratios whose respective correlation coefficient is highest as the selected scalar value.