Abstract: Disclosed in some examples are methods, systems, memory devices, and machine-readable mediums which increase read throughput by introducing a delay prior to issuing a command to increase the chances that read commands can be executed in parallel. Upon receipt of a read command, if there are no other read commands in the command queue for a given portion (e.g., plane or plane group) of the die, the controller can delay issuing the read command for a delay period using a timer. If, during the delay period, an eligible read command is received, the delayed command and the newly received command are both issued in parallel using a multi-plane read. If no eligible read command is received during the delay period, the read command is issued after the delay period expires.

Abstract: A memory module includes a plurality of memory integrated circuit (IC) packages to store data and a command buffer IC to buffer one or more memory commands destined for the memory IC packages. The command buffer IC includes a first interface circuit and one or more second interface circuits. The first interface circuit receives the one or more memory commands. The one or more second interface circuits output a pre-programmed command sequence to one or more devices separate from the command buffer IC, the pre-programmed command sequence output in response to the one or more memory commands matching a pre-programmed reference command pattern.

Abstract: This disclosure relates to high-performance computing, and more particularly to techniques for kernel-assisted device polling of user-space devices. A common kernel-based polling mechanism is provided for concurrently handling both kernel-based polling for kernel-space devices such as network interfaces (e.g., network NICs) and kernel-based polling for user-space devices such as remote direct memory access devices (e.g., RDMA NICs). Embodiments perform kernel-based polling on a first device that has a corresponding device driver in an operating system kernel. Using the same polling mechanism, the kernel-based polling is performed on a second device, the second device being a user-space device wherein the kernel-based polling on the second device is configured by creating a second device file descriptor that is not associated with a corresponding device driver in the operating system kernel.

Abstract: Examples provided herein provide a docking device for a computing device. In one example, the docking device includes a magnet array to align the computing device with a mounting device on the docking device. The docking device includes a locking blade that inserts from the mounting device into the computing device in a 3-dimensional (3D) motion.

Abstract: A method includes receiving, at a first computing device, a first input/output (IO) command from a first artificial intelligence processing unit (AI PU), the first IO command associated with a first AI model training operation. The method further includes receiving, at the first computing device, a second IO command from a second AI PU, the second IO command associated with a second AI model training operation. The method further includes assigning a first timestamp to the first IO command based on a first bandwidth assigned to the first AI model training operation. The method further includes assigning a second timestamp to the second IO command based on a second bandwidth assigned to the second AI model training operation.

Abstract: Described is a system, method, and computer program product for performing elections in a database cluster, where system resource statistics information is used to predict a cluster node failure. Resource statistics data is classified and used to identify anomalies. The anomalies can be used to determine the probability of a cluster node failure and to then elect a new master node and/or surviving sub-cluster.

Abstract: Methods, apparatus, and processor-readable storage media for device component management using deep learning techniques are provided herein. An example computer-implemented method includes obtaining telemetry data from one or more enterprise devices; determining, for each of the one or more enterprise devices, values for multiple device attributes by processing the obtained telemetry data; generating, for each of the one or more enterprise devices, at least one prediction related to lifecycle information of at least one device component by processing the determined attribute values using one or more deep learning techniques; and performing one or more automated actions based at least in part on the at least one generated prediction.

Abstract: A method includes asserting a field of an event flag mask register configured to inhibit an event handler. The method also includes, responsive to an event that corresponds to the field of the event flag mask register being triggered: asserting a field of an event flag register associated with the event; and based the field in the event flag register being asserted, taking an action by a task being executed by the data processor core.

Abstract: The present disclosure relates to a mirrored serial interface (MSI) for accessing peripherals through four wire serial interface. More particularly, the present disclosure is related to serial peripheral protocol with looped back mechanism in which contents of source data line are looped back onto the destination line and compared at every clock edge to ensure data sanity and to assert presence of slave and master device during and between cycles.

Abstract: A process includes controlling a current source to cause the current source to provide a plurality of different currents at different times to an output communication line of a video display interface; and acquiring a plurality of voltages corresponding to the plurality of different currents. Acquiring the plurality of voltages includes sampling a voltage of the output communication line. The process includes comparing the plurality of voltages to a plurality of voltage thresholds; and based on a result of the comparison, determining whether the video display interface is coupled to a cable-based far end termination.

Abstract: A method and apparatus for the increase of internal data throughput and processing capability for SSD's, to enable processing of database commands on an SSD. A front-end ASIC is provided with 256 to 512 RISC processing cores to enable decomposition and parallelization of host commands to front-end module (FM) ASICs that each in turn are coupled to multiple NVM dies, as well as processing of host database operations such as insert, select, update, and delete. Each FM ASIC is architected to increase parity bits to 33.3% of NVM data, and process parity data with 14 LDPC's. By increasing the parity bits to 33.3%, BER is reduced, power consumption is reduced, and data throughput within the SSD is increased.

Abstract: An information processing apparatus according to an aspect of the present invention includes an information processing circuit configured to generate a finite state machine based on a predetermined matching condition with respect to sequence data of an event that is input to the information processing apparatus; to process the sequence data so as to substantially remove data that does not match the matching condition from the sequence data; and to output the processed sequence data.

Abstract: A storage device includes a non-volatile memory including a plurality of blocks, a buffer memory that stores a plurality of on-cell counts, which are generated by reading memory cells connected to a plurality of reference word lines of the plurality of blocks by using a read level, and an artificial neural network model, and a controller that inputs an on-cell count corresponding to a target block among the plurality of on-cell counts and a number of a target word line of the target block to the artificial neural network model, and infers a plurality of read levels for reading data of memory cells connected to the target word line using the artificial neural network model.

Abstract: Technologies are described herein for providing a Baseboard Management Controller (“BMC”) -based security processor. The disclosed BMC-based security processor can provide a hardware Root of Trust (“RoT”) for a computing platform without the addition of specialized silicon to the platform and while minimizing the number of attack points. The disclosed BMC-based security processor can also provide functionality for securely filtering requests made on certain buses in a computing platform. Through implementations of the features identified briefly above, and others described herein, various technical benefits can be achieved such as, but not limited to, increased security as compared to previous computing systems that utilize a BMC to provide a hardware RoT and reduced complexity and cost as compared to previous computing systems that utilize a separate hardware device, such as a Field Programmable Gate Array (“FPGA”) or a microcontroller, to provide a hardware RoT.

Abstract: An embodiment of the invention may include a method, computer program product, and computer system for reconfiguration of a computing environment from an as-is input/output (I/O) configuration to a to-be I/O configuration. An embodiment may include normalizing respective hierarchical models of the as-is and the to-be I/O configurations. The hierarchical models each comprise a hierarchical structure of leaf and non-leaf nodes. Normalizing comprises the application of syntactical transformation rules to the hierarchical models such that their respective I/O configurations are defined in a syntactically consistent manner. An embodiment may include creating respective hash tree representations of the first hierarchical model and the second hierarchical model. Nodes of the hash tree representations are checksum values. Nodes of the hash tree representations reflect the hierarchical structure of their respective hierarchical models.

Abstract: This disclosure describes system on a chip (SOC) communications that prevent direct memory access (DMA) attacks. An example SoC includes an encryption engine and a security processor. The encryption engine is configured to encrypt raw input data using a cipher key to form an encrypted payload. The security processor is configured to select the cipher key from a key store holding a plurality of cipher keys based on a channel ID describing a {source subsystem, destination subsystem} tuple for the encrypted payload, to form an encryption header that includes the channel ID, to encapsulate the encrypted payload with the encryption header that includes the channel ID to form a crypto packet, and to transmit the crypto packet to a destination SoC that is external to the SoC.

Abstract: Disclosed is a data transfer system capable of accelerating data transmission between two chips. The data transfer system includes: a master system-on-a-chip (SoC) including a master transmission circular buffer and a master reception circular buffer; and a slave SoC including a slave reception circular buffer and a slave transmission circular buffer. The slave/master reception circular buffer is a duplicate of the master/slave transmission circular buffer; accordingly, the write pointers of the two corresponding buffers are substantially synchronous and the read pointers of the two corresponding buffers are substantially synchronous as well. In light of the above, the read and write operations of the master/slave transmission circular buffer can be treated as the read and write operations of the slave/master reception circular buffer; therefore some conventional data reproducing procedure(s) for the data transmission can be omitted and the data transmission is accelerated.

Abstract: A neural processing unit (NPU) includes a controller including a scheduler, the controller configured to receive from a compiler a machine code of an artificial neural network (ANN) including a fusion ANN, the machine code including data locality information of the fusion ANN, and receive heterogeneous sensor data from a plurality of sensors corresponding to the fusion ANN; at least one processing element configured to perform fusion operations of the fusion ANN including a convolution operation and at least one special function operation; a special function unit (SFU) configured to perform a special function operation of the fusion ANN; and an on-chip memory configured to store operation data of the fusion ANN, wherein the schedular is configured to control the at least one processing element and the on-chip memory such that all operations of the fusion ANN are processed in a predetermined sequence according to the data locality information.

Abstract: A computer system is provided. The computer system includes a memory and at least one processor coupled to the memory. The processor is configured to identify a message to a plug and play (PnP) manager of an operating system, the message comprising an identifier of a device to be configured by the PnP manager, determine whether the device is targeted for device identifier translation at least in part by determining whether the device satisfies one or more target device criteria, and replace the identifier of the device with a reference identifier different from the identifier of the device in response to a determination that the device is targeted for device identifier translation, the reference identifier being usable by the PnP manager to install or configure the device.

Abstract: The present invention includes a CP monitoring unit configured to monitor a CP voltage applied through a CP line from the controller of the slow charging cable when the slow charging cable is connected to an inlet, a mode switching request unit configured to request mode switching to the controller of the slow charging cable by converting the monitored CP voltage to a preset mode switching voltage; a communication switching unit configured to connect a LIN transceiver for LIN communication to the CP line when the controller of the slow charging cable is switched to a LIN communication mode for software update, and a control unit configured to control the LIN transceiver to transmit a pre-stored software update file to the controller of the slow charging cable, so that the controller's software can be updated without disassembling or damaging the controller of the slow charging cable.