A memory system includes a memory array having a plurality of memory cells; and a controller coupled to the memory array, the controller configured to: designate a storage mode for a target set of memory cells based on valid data in a source block, wherein the target set of memory cells are configured with a capacity to store up to a maximum number of bits per cell, and the storage mode is for dynamically configuring the target set of memory cells in as cache memory that stores a number of bits less per cell than the corresponding maximum capacity.

The present invention provides a control method of a flash memory controller, wherein the flash memory controller is configured to access a flash memory module, and the control method includes the steps of: receiving a settling command from a host device to configure a portion space of the flash memory module as a zoned namespace; receiving a write command from the host device to write data corresponding a first zone into a plurality of blocks of the flash memory module, wherein an access mode chose by the flash memory controller is determined based on a size of each zone and a size of each block.

Technologies for addressing individual bits in memory include a device having a memory that includes partitions that each have tiles, in which each tile stores an individual bit. The device also includes circuitry to receive a request to access (e.g., read or write) a sequence of bits in a partition. The request specifies a logical row or column address. A corresponding tile is determined from the logical row or column address and for each bit in the sequence. The corresponding tile is accessed to read or write the bit therein.

A method, non-transitory computer readable medium, and device that assists with reducing memory fragmentation in solid state devices includes identifying an allocation area within an address range to write data from a cache. Next, the identified allocation area is determined for including previously stored data. The previously stored data is read from the identified allocation area when it is determined that the identified allocation area comprises previously stored data. Next, both the write data from the cache and the read previously stored data are written back into the identified allocation area sequentially through the address range.

An apparatus to facilitate memory data compression is disclosed. The apparatus includes a memory and having a plurality of banks to store main data and metadata associated with the main data and a memory management unit (MMU) coupled to the plurality of banks to perform a hash function to compute indices into virtual address locations in memory for the main data and the metadata and adjust the metadata virtual address locations to store each adjusted metadata virtual address location in a bank storing the associated main data.

A technique relates to flash-optimized data layout of a dataset for queries. Selection columns are stored in flash memory according to a selection optimized layout, where the selection optimized layout is configured to optimize predicate matching and data skipping. The selection optimized layout, for each selection column, is formed by storing a selection column dictionary filled with unique data values in a given selection column, where the unique data values are stored in sorted order in the selection column dictionary. Row position designations are stored corresponding to each row position that the unique data values are present within the given selection column, without duplicating storage of any of the unique data values that occur more than once in the given selection column.

According to the present disclosure is provided a device and method for mapping management in a flash memory based on partitioning the memory to a main address space and a substitute space, each partition comprising locations in the memory that are denoted by at least three statues according to which locations are mapped from the main space to the substitute space while responsively modifying the statuses.

The present invention relates to a dynamic memory management method which includes generating an N-dimensional memory address space in which coordinates are in a range of N natural numbers, the sum of which is the number of bits; and mapping a predetermined linear memory address region to an address region in the N-dimensional memory address space.

A memory sub-system configured to dynamically determine input/output sizes of write commands based on a media physical layout of a memory sub-system. The memory sub-system can identify, dynamically in response to write commands being selected for execution in media units of the memory sub-system, a portion of a media layout that maps from logical addresses identified by the write commands in the logical address space to physical addresses of memory units in the media units. Based on the media layout, an input/output size for a next write command is identified and transmitted to the host system in a response. The host system generates the next write command and configures the amount of data to be written through the next write command based on based on the input/output size identified in the response.

Described herein is a graphics processing unit (GPU) comprising a first processing cluster to perform parallel processing operations, the parallel processing operations including a ray tracing operation and a matrix multiply operation; and a second processing cluster coupled to the first processing cluster, wherein the first processing cluster includes a floating-point unit to perform floating point operations, the floating-point unit is configured to process an instruction using a bfloat16 (BF16) format with a multiplier to multiply second and third source operands while an accumulator adds a first source operand with output from the multiplier.