Abstract: A chip or integrated circuit includes a layer that includes a first device and a second device. A scribe line is located between the first device and the second device and separates the first device from the second device. An electrically conductive connection traverses the scribe line and is coupled to the first device and the second device, thus connecting the first and second devices.

Abstract: Embodiments of the disclosure provide systems and methods for allocating memory space in a memory device. The system can include: a memory device for providing the memory space; and a compiler component configured for: receiving a request for allocating a data array having a plurality of data elements in the memory device, wherein each of the plurality of data elements has a logical address; generating an instruction for allocating memory space for the data array in the memory device based on the request; generating device addresses for the plurality of data elements in the memory device based on logical addresses of the plurality of data elements; and allocating the memory space for the data array in the memory device based on the device addresses and the instruction.

Abstract: A 3D-stacked memory device including: a base die including a plurality of switches to direct data flow and a plurality of arithmetic logic units (ALUs) to compute data; a plurality of memory dies stacked on the base die; and an interface to transfer signals to control the base die.

Abstract: A cache coherency mode includes: in response to a read request from a device in the host-device system for an instance of the shared data, sending the instance of the shared data from the host device to that device; and, in response to write request from a device, storing data associated with the write request in the cache of the host device. Shared data is pinned in the cache of the host device, and is not cached in any of the other devices in the host-device system. Because there is only one cached copy of the shared data in the host-device system, the devices in that system are cache coherent.

Abstract: A high bandwidth memory (HBM) system includes a first HBM+ card. The first HBM+ card includes a plurality of HBM+ cubes. Each HBM+ cube has a logic die and a memory die. The first HBM+ card also includes a HBM+ card controller coupled to each of the plurality of HBM+ cubes and configured to interface with a host, a pin connection configured to connect to the host, and a fabric connection configured to connect to at least one HBM+ card.

Abstract: A method of correcting a memory error of a dynamic random-access memory module (DRAM) using a double data rate (DDR) interface, the method includes conducting a memory transaction including multiple bursts with a memory controller to send data from data chips of the DRAM to the memory controller, detecting one or more errors using an ECC chip of the DRAM, determining a number of the bursts having the errors using the ECC chip of the DRAM, determining whether the number of the bursts having the errors is greater than a threshold number, determining a type of the errors, and directing the memory controller based on the determined type of the errors, wherein the DRAM includes a single ECC chip per memory channel.

Abstract: A general matrix-matrix multiplication (GEMM) dataflow accelerator circuit is disclosed that includes a smart 3D stacking DRAM architecture. The accelerator circuit includes a memory bank, a peripheral lookup table stored in the memory bank, and a first vector buffer to store a first vector that is used as a row address into the lookup table. The circuit includes a second vector buffer to store a second vector that is used as a column address into the lookup table, and lookup table buffers to receive and store lookup table entries from the lookup table. The circuit further includes adders to sum the first product and a second product, and an output buffer to store the sum. The lookup table buffers determine a product of the first vector and the second vector without performing a multiply operation. The embodiments include a hierarchical lookup architecture to reduce latency. Accumulation results are propagated in a systolic manner.

Abstract: Embodiments of the disclosure provide methods and systems for memory management. The method can include: receiving a request for allocating target node data to a memory space, wherein the memory space includes a buffer and an external memory and the target node data comprises property data and structural data and represents a target node of a graph having a plurality of nodes and edges; determining a node degree associated with the target node data; allocating the target node data to the memory space based on the determined node degree.

Abstract: Embodiments of the disclosure provide memory devices and methods related to memory accessing.

Abstract: A 3D-stacked memory device including: a base die including a plurality of switches to direct data flow and a plurality of arithmetic logic units (ALUs) to compute data; a plurality of memory dies stacked on the base die; and an interface to transfer signals to control the base die.

Abstract: Methods and systems are provided for improving memory control. A memory architecture includes a plurality of memory units and an interface. A respective memory unit of the plurality of memory units is configured with a Processing-In-Memory (PIM) architecture. The interface includes a plurality of lines. The interface is coupled between the plurality of memory units and a host. The interface is configured to receive one or more signals from a host via the plurality of lines. The respective memory unit of the plurality of memory units is coupled with a respective line of the plurality of lines, and the respective memory unit is further configured to receive a respective signal of the one or more signals via the interface so as to be individually selected by the host.

Abstract: A memory module includes a random access memory (RAM) device that includes a first storage region and a second storage region, a nonvolatile memory device, and a controller that controls the RAM device or the nonvolatile memory device under control of a host. The controller includes a data buffer that temporarily stores first data received from the host, and a buffer returning unit that transmits first release information to the host when the first data are moved from the data buffer to the first storage region or the second storage region of the RAM device and transmits second release information to the host when the first data are moved from the second storage region to the nonvolatile memory device.

Abstract: Memory control methods and systems are provided. A memory architecture includes one or more accelerators, a controller, and a transactional interface. A respective accelerator of the one or more accelerators includes a respective storage area configured to store data and a respective computation unit configured to perform computation. The respective storage area and the respective computation unit are configured to interact with each other. The controller is coupled with the one or more accelerators. The controller is configured to control the one or more accelerators, receive a command from a host, and perform an operation in response to receiving the command. The transactional interface is coupled between the controller and the host and includes a command and address signal channel, which is configured to transfer command and address signals from the host to the controller.

Abstract: A method of page size aware scheduling and a non-transitory computer-readable storage medium having recorded thereon a computer program for executing the method of page size aware scheduling are provided. The method includes determining a size of a media page; determining if the media page is open or closed; performing, by a memory controller, a speculative read operation if the media page is determined to be open; and performing, by the memory controller, a regular read operation if the media page is determined to be closed.

Abstract: Methods and systems are provided for improving memory control. A memory architecture includes a plurality of memory units and an interface. A respective memory unit of the plurality of memory units is configured with a Processing-In-Memory (PIM) architecture. The interface includes a plurality of lines. The interface is coupled between the plurality of memory units and a host. The interface is configured to receive one or more signals from a host via the plurality of lines. The respective memory unit of the plurality of memory units is coupled with a respective line of the plurality of lines, and the respective memory unit is further configured to receive a respective signal of the one or more signals via the interface so as to be individually selected by the host.

Abstract: A method of correcting a memory error of a dynamic random-access memory module (DRAM) using a double data rate (DDR) interface, the method includes conducting a memory transaction including multiple bursts with a memory controller to send data from data chips of the DRAM to the memory controller, detecting one or more errors using an ECC chip of the DRAM, determining a number of the bursts having the errors using the ECC chip of the DRAM, determining whether the number of the bursts having the errors is greater than a threshold number, determining a type of the errors, and directing the memory controller based on the determined type of the errors, wherein the DRAM includes a single ECC chip per memory channel.

Abstract: A high bandwidth memory system. In some embodiments, the system includes: a memory stack having a plurality of memory dies and eight 128-bit channels; and a logic die, the memory dies being stacked on, and connected to, the logic die; wherein the logic die may be configured to operate a first channel of the 128-bit channels in: a first mode, in which a first 64 bits operate in pseudo-channel mode, and a second 64 bits operate as two 32-bit fine-grain channels, or a second mode, in which the first 64 bits operate as two 32-bit fine-grain channels, and the second 64 bits operate as two 32-bit fine-grain channels.

Abstract: A method of correcting a memory error of a dynamic random-access memory module (DRAM) using a double data rate (DDR) interface, the method includes conducting a memory transaction including multiple bursts with a memory controller to send data from data chips of the DRAM to the memory controller, detecting one or more errors using an ECC chip of the DRAM, determining a number of the bursts having the errors using the ECC chip of the DRAM, determining whether the number of the bursts having the errors is greater than a threshold number, determining a type of the errors, and directing the memory controller based on the determined type of the errors, wherein the DRAM includes a single ECC chip per memory channel.

Abstract: The present disclosure provides devices and methods for refreshing memory. The method includes receiving information of a memory array in an accelerator coupled to a host device, and when the accelerator operates in a first refresh mode, based on the received information: disabling one or more refresh operations of a memory controller; and refreshing one or more rows of the memory array by the host device.

Abstract: A method of processing in-memory commands in a high-bandwidth memory (HBM) system includes sending a function-in-HBM instruction to the HBM by a HBM memory controller of a GPU. A logic component of the HBM receives the FIM instruction and coordinates the instructions execution using the controller, an ALU, and a SRAM located on the logic component.