Abstract: Embodiments are directed towards apparatuses, methods, and systems including a pre-read command to eliminate an additional access of read data from a storage location of a memory device. In embodiments, a memory controller issues a pre-read command to store read data in a pre-read latch. In embodiments, the command is issued during a first access of the read data from a storage location in connection with a modify-write operation of the read data. In embodiments, the pre-read latch is located in or coupled to a selected partition of a memory device that includes the storage location that stores the read data. In embodiments, the memory controller subsequently issues a modify-write command to compare the read data stored in the pre-read latch with incoming data, to eliminate a need for a second access of the storage location during completion of the modify-write operation. Additional embodiments may be described and claimed.

Abstract: Technologies for providing a scalable architecture to efficiently perform compute operations in memory include a memory having media access circuitry coupled to a memory media. The media access circuitry is to access data from the memory media to perform a requested operation, perform, with each of multiple compute logic units included in the media access circuitry, the requested operation concurrently on the accessed data, and write, to the memory media, resultant data produced from execution of the requested operation.

Abstract: Examples are disclosed for adaptive configuration of non-volatile memory. The examples include a mode register configured to include default and updated values to indicate one or more configurations of the non-volatile memory. The examples may also include discoverable capabilities maintained in a configuration table that may indicate memory address lengths and/or operating power states.

Abstract: Uncorrectable memory errors may be reduced by determining a logical array address for a set of memory arrays and transforming the logical array address to at least two unique array addresses based, at least in part, on logical locations of at least two memory arrays within the set of memory arrays. The at least two memory arrays are then accessed using the at least two unique array addresses, respectively.

Abstract: A memory device performs DLL (delay locked loop) calibration in accordance with a DLL calibration mode configured for the memory device. A host controller can configure the calibration mode based on operating conditions for the memory device. The memory device includes an input/output (I/O) interface circuit and a delay locked loop (DLL) circuit coupled to control I/O timing of the I/O interface. A control circuit of the memory device selectively enables and disables DLL calibration in accordance with the DLL calibration mode. When selectively enabled, the DLL calibration is to operate at a time interval identified by the DLL calibration mode, and when selectively disabled, the DLL calibration is to cease or refrain from DLL calibration operations.

Abstract: Cross point memory architectures, devices, systems, and methods are disclosed and described, and can include a cross point memory core subsystem having increased bandwidth that is scalable. The memory core can include a plurality of independently operating partitions, each comprising a plurality of cross point memory arrays.

Abstract: Technologies for provisioning error-corrected data for use in in-memory compute operations include a memory that includes a memory media having multiple memory partitions and media access circuitry coupled to the memory media. The media access circuitry is to receive a request to perform an in-memory compute operation on data from the memory media. The request specifies a memory partition of the memory media in which the data is located. The media access circuitry reads the data from the memory partition. The media access circuitry performs error correction on the read data to produce error-corrected read data and stores the error-corrected read data in a temporary buffer for access by one or more in-memory compute operations, in addition to the requested in-memory compute operation.

Abstract: According to one embodiment, an apparatus is disclosed. The apparatus includes a memory device having a device identification. The apparatus further includes a low power wake circuit configured to receive a low power wake signal and an identification information, and further configured to initiate a transition of the memory device from a low power state to an active state responsive to an active low power wake signal and the wake identification information matching the device identifier.

Abstract: Technologies for preserving error correction capability in compute-near-memory operations in a memory include memory media and a media access circuitry coupled with the memory media. The media access circuitry is to detect an error code adjustment state indicative of a failure in the initiated error correction. The media access circuitry is to adjust a voltage to the memory media to eliminate the error code correction adjustment state. Once eliminated, the media access circuitry is to perform the error correction on the read data.

Abstract: Technologies for providing multiple tier memory media management include a memory having a media access circuitry connected to a memory media. The media access circuitry is to receive a request to perform an in-memory compute operation. Additionally, the media access circuitry is to read, in response to the request, data from a memory media region of the memory media, write the read data into a compute media region of the memory, perform, on the data in the compute media region, the in-memory compute operation, write, to the memory media region, resultant data indicative of a result of performance of the in-memory compute operation.

Abstract: Technologies for providing adaptive memory media management include media access circuitry connected to a memory media. The media access circuitry is to receive a request to perform at least one memory access operation to be managed by the media access circuitry. The media access circuitry is further to manage the requested at least one memory access operation, including disabling a memory controller in communication with the media access circuitry from managing the memory media while the at least one requested memory access operation is performed.

Abstract: Technologies for performing in-memory macro operations include a memory having a media access circuitry connected to a memory media. The media access circuitry is to receive a request to perform an in-memory macro operation indicative of a set of multiple in-memory operations. The media access circuitry is also to perform, in response to the request, the in-memory macro operation on data present in the memory media.

Abstract: A memory device performs DLL (delay locked loop) calibration in accordance with a DLL calibration mode configured for the memory device. A host controller can configure the calibration mode based on operating conditions for the memory device. The memory device includes an input/output (I/O) interface circuit and a delay locked loop (DLL) circuit coupled to control I/O timing of the I/O interface. A control circuit of the memory device selectively enables and disables DLL calibration in accordance with the DLL calibration mode. When selectively enabled, the DLL calibration is to operate at a time interval identified by the DLL calibration mode, and when selectively disabled, the DLL calibration is to cease or refrain from DLL calibration operations.

Abstract: Technologies for performing tensor operations in memory include a memory comprising media access circuitry coupled to a memory media having a cross point architecture. The media access circuitry is to access matrix data from the memory media, perform a tensor operation on the matrix data, and write, to the memory media, resultant data indicative of a result of the tensor operation.

Abstract: Technologies for providing high efficiency compute architecture on cross point memory for artificial intelligence operations include a memory that includes media access circuitry coupled to a memory media having a cross point architecture. The media access circuitry is to access matrix data from the memory media, including broadcasting matrix data associated with one partition of the memory media to multiple other partitions of the memory media. The media access circuitry is also to perform, with each of multiple compute logic units associated with different partitions of the memory media, a tensor operation on the matrix data and write, to the memory media, resultant data indicative of a result of the tensor operation.

Abstract: Technologies for providing a scalable architecture to efficiently perform compute operations in memory include a memory having media access circuitry coupled to a memory media. The media access circuitry is to access data from the memory media to perform a requested operation, perform, with each of multiple compute logic units included in the media access circuitry, the requested operation concurrently on the accessed data, and write, to the memory media, resultant data produced from execution of the requested operation.

Abstract: Technologies for providing multiple levels of error correction include a memory that includes media access circuitry coupled to a memory media. The media access circuitry is to read data from the memory media. Additionally, the media access circuitry is to perform, with an error correction logic unit located in the media access circuitry, error correction on the read data to produce error-corrected data.

Abstract: Cross point memory architectures, devices, systems, and methods are disclosed and described, and can include a cross point memory core subsystem having increased bandwidth that is scalable. The memory core can include a plurality of independently operating partitions, each comprising a plurality of cross point memory arrays.

Abstract: Uncorrectable memory errors may be reduced by determining a logical array address for a set of memory arrays and transforming the logical array address to at least two unique array addresses based, at least in part, on logical locations of at least two memory arrays within the set of memory arrays. The at least two memory arrays are then accessed using the at least two unique array addresses, respectively.

Abstract: Several systems and methods of chip select are described. In one such method, a device maintains two identifiers, (ID_a and ID_m). When the device receives a command, it examines the values of ID_a and ID_m relative to a third reference identifier (ID_s). If either ID_a or ID_m is equivalent to ID_s, the device executes the command, otherwise, the device ignores the command. By using two different identification methods, a system has options in choosing to activate devices, being able to selectively switch between selecting multiple devices and single devices in a quick manner. In another such method, a device may have a persistent area that stores identification information such as an ID_a. Thus, system functionality may remain independent from any defect/marginality associated with the physical or logical components required for initial ID_a assignment of all devices in the system.