A method is for operating a nonvolatile dual in-line memory module (NVDIMM). The NVDIMM includes a dynamic random access memory (DRAM) and a nonvolatile memory (NVM) device, the DRAM including a first input/output (I/O) port and a second I/O port, and the second I/O port connected to the NVM device. The method includes receiving an externally supplied command signal denoting a read/write command and a transfer mode, driving a multiplexer to select at least one of the first and second I/O ports according to the transfer mode of the command signal, and reading or writing data according to the read/write command of the command signal in at least one of the DRAM and NVM device using the at least one of the first and second I/O ports selected by driving the multiplexer.

Systems and methods are provided for implementing hardware optimization for a hardware accelerator. The hardware accelerator emulates a neural network. Training of the neural network integrates a regularized pruning technique to systematically reduce a number of weights. A crossbar array included in hardware accelerator can be programmed to calculate node values of the pruned neural network to selectively reduce the number of weight column lines in the crossbar array. During deployment, the hardware accelerator can be programmed to power off periphery circuit elements that correspond to a pruned weight column line to optimize the hardware accelerator for power. Alternatively, before deployment, the hardware accelerator can be optimized for area by including a finite number of weight column line. Then, regularized pruning of the neural network selectively reduces the number of weights for consistency with the finite number of weight columns lines in the hardware accelerator.

A memory device includes a memory array of memory cells, wordlines and bitlines connected to the memory cells, a first read multiplexor and a second read multiplexor connected to the bitlines, a first sense amplifier connected to the first read multiplexor, a second sense amplifier connected to the second read multiplexor, a first data path connected to the first sense amplifier, and a second data path connected to the second sense amplifier. Each of the memory cells is connected to only one pair of the bitlines and only one of the wordlines. The first read multiplexor is adapted to connect the first sense amplifier to the bitlines during a first portion of a clock cycle and the second read multiplexor is adapted to connect the second sense amplifier to the bitlines during a second portion of a clock cycle that is different from the first portion of the clock cycle.

A memory includes S storage blocks, N global bitlines, and a signal amplification circuit. Each of the S storage blocks is connected to the N global bitlines, the N global bitlines are connected to the signal amplification circuit, the signal amplification circuit is configured to amplify electrical signals on the N global bitlines, and each storage block includes N columns of storage units, N local bitlines, and N bitline switches. In each storage block, storage units in an i.sup.th column are connected to an i.sup.th local bitline, the i.sup.th local bitline is connected to an i.sup.th global bitline by using an i.sup.th bitline switch in the N bitline switches. A memory array is fine-grained, so that i.sup.th local bitlines in the S storage blocks can share one global bitline.

Methods, systems, and devices for programmable column access are described. A device may transfer voltages from memory cells of a row in a memory array to respective digit lines for the memory cells. The voltages may be indicative of logic values stored at the memory cells. The device may communicate respective control signals to a set of multiplexers coupled with the digit lines, where each multiplexer is coupled with a respective subset of the digit lines. Each multiplexer may couple a digit line of the respective subset of digit lines with a respective sense component for that multiplexer based on the respective control signal for that multiplexer.

A memory device comprising a cell field having memory cells, N bit lines, which are respectively connected to at least one of the memory cells of the cell field, N being a whole number greater than one, N sense amplifiers; a bit shift circuit, which has S switch element rows, S being a whole number greater than one and a row number in the range from zero to S−1 being assignable to each switch element row. Each switch element row includes at least one semiconductor switch element connected to one of the bit lines and one of the sense amplifiers. Switch elements of each row connect all bit lines, whose bit line number is smaller than or equal to N minus the row number, to sense amplifiers, so that the respective sense amplifier number is equal to the respective bit line number plus the row number.

A memory programming circuit for programming a non-volatile memory device having an array structure includes a plurality of rows, each row having a row index and comprising one or more memory units, each memory unit being configured to receive one or more input signals and to deliver one or more output signals, the memory programming circuit comprising: a first source line connected to the top electrode of the memory units comprised at rows of odd row indices, and a second source line connected to the top electrodes of the memory units comprised at rows of even row indices.

In a method of testing a nonvolatile memory device including a first semiconductor layer in which and a second semiconductor layer is formed prior to the first semiconductor layer, circuit elements including a page buffer circuit are provided in the second semiconductor layer, an on state of nonvolatile memory cells which are not connected to the page buffer circuit is mimicked by providing a conducting path between an internal node of a bit-line connection circuit connected between a sensing node and a bit-line node of the page buffer circuit and a voltage terminal to receive a first voltage, a sensing and latching operation with the on state being mimicked is performed in the page buffer circuit and a determination is made as to whether the page buffer circuit operates normally is made based on a result of the sensing and latching operation.

A computing storage architecture is disclosed. Memory devices may incorporate distributed processors and memory. The devices can be arranged using multiple packages, each package including one, or multiple, dies. In one aspect of the disclosure, any of the processors on a first die may transfer data to and from any processor on a second die internally within the device without having to pass through an external storage controller. In another aspect of the disclosure, a multi-package processing architecture allows for both in-package and inter-channel data transfers between processors within the same device. In still another aspect of the disclosure, one or more processors may include a preemptive scheduler circuit, which enables a processor to interrupt an ongoing lower priority transmission and to immediately transfer data.

A static random access memory (SRAM) includes fast SRAM bit cells and fast multiplexer circuits that are formed in a first row of fast cells in a hybrid standard cell architecture. Slow SRAM bit cells and slow multiplexer circuits are formed in a second row of slow cells. The slow multiplexer circuits provide a column output for the fast SRAM bit cells and the fast multiplexer circuits provide a column output for the slow SRAM bit cells. Thus, one SRAM column has fast bit cells and slow multiplexer stages while the adjacent SRAM column has slow bit cells and fast multiplexer stages to thereby provide an improved performance balance when reading the SRAM.