Various systems and methods for computer memory overcommitment management are described herein. A system for computer memory management includes a memory device to store data and a mapping table; and a memory overcommitment circuitry to: receive a signal to move data in a first block from a memory reduction area in the memory device to a non-memory reduction area in the memory device, the memory reduction area to store data using a memory reduction technique, and the non-memory reduction area to store data without any memory reduction techniques; allocate a second block in the non-memory reduction area; copy the data in the first block to the second block; and update the mapping table to revise a pointer to point to the second block, the mapping table used to store pointers to memory device in the memory reduction area and the non-memory reduction area.

A NOT logic circuit is provided comprising: one or more memory devices; wherein a first memory address location of the one or more memory devices stores first content data, wherein the first content data includes a first ternary value and a corresponding first priority value, wherein the first ternary value includes a continuous sequence of X-state values that represent a first range of non-X ternary values; wherein a second memory address of the one or more memory device stores second content data that includes a second ternary value and a corresponding second priority value, wherein the second ternary value includes a continuous sequence of non-X state values represent a non-X ternary value that is within the first range of non-X ternary values; an interface is coupled to receive a ternary value from a processing device; comparator circuitry operable to compare a received ternary key with the outputted first ternary value and to compare the received ternary key with the outputted second ternary value; priority encoder logic operable to, return the outputted first priority value on a condition that the received ternary key matches the first ternary value and the received ternary key does not match second ternary value, and return the outputted second priority value on a condition that the received ternary key matches the first ternary value and that the received ternary key matches the second ternary value; and detection logic operable to send a return to the processing device on a condition of a return of the first priority value.

A system, a computer-implemented method, and a computer readable medium having stored thereon a computer executable program code for updating a database stored across a set of partitions on a computer system. The computer system comprises a first and a second storage device and a computing device. The set of partitions comprises a first subset of partitions and a second subset of partitions. The first storage device stores the first subset of partitions. The second storage device stores the second subset of partitions. The computing device comprises a memory storing processor-executable program code and a computer processor to execute the processor-executable program code in order to cause the computing device to execute the computer-implemented method.

Technologies for composing a managed node with multiple processors on multiple compute sleds to cooperatively execute a workload include a memory, one or more processors connected to the memory, and an accelerator. The accelerator further includes a coherence logic unit that is configured to receive a node configuration request to execute a workload. The node configuration request identifies the compute sled and a second compute sled to be included in a managed node. The coherence logic unit is further configured to modify a portion of local working data associated with the workload on the compute sled in the memory with the one or more processors of the compute sled, determine coherence data indicative of the modification made by the one or more processors of the compute sled to the local working data in the memory, and send the coherence data to the second compute sled of the managed node.

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.

Provided are a computer program product, system, and method for dynamically increasing capacity of a storage device. For address mappings, each addressing mapping indicates a storage device block address for a host block address and a compressed block size indicating a number of blocks storing compressed data for data written to the host block address starting at the storage device block address. Write data for a write request to a host block address is compressed to produce compressed data. A block size of the compressed data is less than request block size of the write data for the write request. Indication is made in the address mapping for the host block address of a storage device address at which to start storing the compressed data in the storage device and the compressed block size. The compressed data is sent to the storage device to write at the storage device block address.

The system contains at least one basic block formed by a first multiplexer having an output is connected to a flag register memory, implemented as a LUT table. An output of a circuit for write permit to the memory is connected to the input of the write signal to the memory, which is further equipped with the clock signal input and the data input. The data output from the memory of each basic block is connected to a masking block relevant for the given basic block. The outputs of these masking blocks are connected to the inputs of the second multiplexer, while its output is the output of the system of flags. The input of the control signal for writing to the memory of each basic block is connected to the output of the demultiplexer and to the second input of the masking block for the given basic block.

Devices and techniques for host accelerated operations in managed NAND devices are described herein. A host logical-to-physical (L2P) table of the NAND device has an associated map. Entries in the map correspond to one or more logical addresses (LA) and indicate whether the host L2P table is current for those LAs. If the table is not current, then a request will bypass the host L2P table, using a standard device L2P lookup instead. Otherwise, the host L2P table can be used.

Methods, devices, and systems for prefetching data. First data is loaded from a first memory location. The first data in cached in a cache memory. Other data is prefetched to the cache memory based on a compression of the first data and a compression of the other data. In some implementations, the compression of the first data and the compression of the other data are determined based on metadata associated with the first data and metadata associated with the other data. In some implementations, the other data is prefetched to the cache memory based on a total of a compressed size of the first data and a compressed size of the other data being less than a threshold size. In some implementations, the other data is not prefetched to the cache memory based on the other data being uncompressed.

The present disclosure generally relates to more efficient use of a delta buffer. To utilize the delta buffer, an efficiency can be gained by utilizing absolute delta entries and relative delta entries. The absolute delta entry will include the type of delta entry, the L2P table index, the L2P table offset, and the PBA. The relative delta entry will include the type of delta entry, the L2P table offset, and the PBA offset. The relative delta entry will utilize about half of the storage space of the absolute delta entry. The relative delta entry can be used after an absolute delta entry so long as the relative delta entry is for data stored in the same block as the previous delta entry. If data is stored in a different block, then the delta entry will be an absolute delta entry.