Abstract: Methods are provided for tactically deploying machine learning operations within existing storage devices without additional capital investment. Machine learning operations can be processed within a SoC of a storage device as embedded software. Storage device designed to utilize machine learning methods within existing configurations can include a non-volatile memory for storing data and executable instructions and a processor to conduct a variety of steps. The steps can include executing a plurality of applications stored in the non-volatile memory, and receiving a request for data, including measurements, from at least one of the plurality of applications. The steps can further determine if the requested data is suitable for substitution by an inference and subsequently select at least one machine learning model for generating a suitable inference.

Abstract: The present disclosure generally relates to data storage devices, such as solid state drives. A data storage device includes a controller, one or more volatile memory locations, and one or more non-volatile memory locations. Computations, including reinforcement learning algorithms, may be completed by the controller using the one or more non-volatile memory locations. Data associated with reinforcement learning is stored in a table on one or more planes of the non-volatile memory, where the results from the computations update the table with the relevant values. The data in the table are aligned to one or more wordlines, such that sensing the wordline senses all the data stored in the table.

Abstract: A processor instantiates a virtual network function (VNF) and a probe to monitor at least one metric associated with the VNF. The processor also allocates a pool of ports to the probe. A transceiver establishes one or more first interfaces between the probe and one or more applications using one or more first ports from the pool of ports. Information such as metrics generated by the ports is concurrently exchange between the probe and the applications using the first interfaces. In some cases, a second interface is established between the probe and a monitoring server. The probe reports mission critical events to the applications via the first interfaces and non-mission critical events to the monitoring server via the second interface concurrently with reporting the mission critical events.

Abstract: A method of temperature compensation to read a flash memory device includes determining a state of the flash memory device. An action is selected with a maximum Q-value from a Q-table for the current state during exploitation. A read operation of a code word from the flash memory device is conducted using one or more parameters according to the selected action. The code word is decoded with an error correction code (ECC) process.

Abstract: A compact cold press grinder that fits within a user's hand that includes a cold press grinder and a container that holds consumables, such as fruits, vegetables, seeds, aquatic plants, and/or other nutritious consumable. The cold press grinder crushes and grinds the consumables to a liquid paste or juice using friction forces, shear forces and pressure (e.g. hydraulic press). The cold press grinder includes a housing that is removably attachable to the container. The cold press grinder may include an input for receiving the consumables within the container and a plurality of nip roller pairs that grind the consumables. The cold press grinder has one output to output the consumable as is or the liquid paste or the juice prepared by the cold press grinder and does not include an output to output consumable waste products.

Abstract: Exemplary methods and apparatus are provided for implementing a deep learning accelerator (DLA) or other neural network components within the die of a non-volatile memory (NVM) apparatus using, for example, under-the-array circuit components within the die. Some aspects disclosed herein relate to configuring the under-the-array components to implement feedforward DLA operations. Other aspects relate to backpropagation operations. Still other aspects relate to using an NAND-based on-chip copy with update function to facilitate updating synaptic weights of a neural network stored on a die. Other aspects disclosed herein relate to configuring a solid state device (SSD) controller for use with the NVM. In some aspects, the SSD controller includes flash translation layer (FTL) tables configured specifically for use with neural network data stored in the NVM.

Abstract: A method, system and product, configured to perform: during an execution of a program, obtaining boundaries of a stack frame of a function that is currently present in a stack, wherein said obtaining the boundaries comprises: obtaining a return address of the function in the stack; determining a length of the function using a mapping of return addresses of one or more functions in the program and corresponding lengths of the one or more functions; and determining the boundaries of the stack frame of the function based on a value of a stack pointer of the stack and based on the length of the function; based on the boundaries of the stack frame of the function, determining that the stack frame is overflown; and in response to said determining that the stack frame is overflown, performing a responsive action.

Abstract: A method of assisting transducer placements on a subject's body for applying tumor treating fields includes: determining, based on one or more images associated with a portion of a subject's body, a first image data, wherein the first image data comprises one or more recommended transducer placement positions; determining, based on the one or more images, a second image data, wherein the second image data comprises one or more transducer placement positions; registering the second image data to the first image data; and generating a composite data comprising the one or more transducer placement positions and the one or more recommended transducer array placement positions.

Abstract: Exemplary methods and apparatus are provided for implementing a deep learning accelerator (DLA) or other neural network components within the die of a non-volatile memory (NVM) apparatus using, for example, under-the-array circuit components within the die. Some aspects disclosed herein relate to configuring the under-the-array components to implement feedforward DLA operations. Other aspects relate to backpropagation operations. Still other aspects relate to using an NAND-based on-chip copy with update function to facilitate updating synaptic weights of a neural network stored on a die. Other aspects disclosed herein relate to configuring a solid state device (SSD) controller for use with the NVM. In some aspects, the SSD controller includes flash translation layer (FTL) tables configured specifically for use with neural network data stored in the NVM.

Abstract: A prokaryotic expression system or a recombinant adenovirus system is used to highly efficiently express VZV envelope gE glycoprotein and the flagellin fusion protein thereof. The produced recombinant gE protein, gE flagellin fusion protein and recombinant adenovirus vector, or composition thereof is used to immunize a mouse so as to promote the body to generate gE and VZV-specific antibody titer, as well as gE-specific and VZV-specific cell immunity.

Abstract: A method, system and product for command injection identification. An input hook function is configured to be executed in response to a potential input provisioning event. The input hook function is configured to perform: analyzing a potential input of the potential input provisioning event to identify whether the potential input comprises a command separator and an executable product; and in response to identifying the command separator and the executable product, recording a suspicious input event indicating the command separator and the executable product. An execution hook function is configured to be executed in response to a potential execution event. The execution hook function is configured to perform: in response to a determination that an execution command of the potential execution event comprises the command separator and the executable product of the suspicious input event, flagging the execution command as a command injection attack.

Abstract: Methods, systems, and assemblies for cooling an electronic component are disclosed. A heat sink assembly includes first and second substrates. The first substrate is in thermal contact with the electronic component. A primary channel is formed in the second surface of the first substrate. The primary channel is configured to direct cooling fluid for cooling the electronic component. An array of primary cooling fluid fins is positioned within the primary channel. The array of primary cooling fluid fins includes upstream solid fins and downstream open fins each having an upstream opening and downstream sidewalls. The secondary channel is formed within the second surface of the second substrate and is configured to direct partially heated cooling fluid away from the electronic component. An array of secondary cooling fluid fins is positioned within the secondary channel downstream. An enclosing layer seals the secondary channel.

Abstract: Disclosed herein are memory devices, systems, and methods of encoding and decoding data. In one aspect, an encoded data chunk is received and segmented into data segments with similar features. Each segment can be decoded based on its features. Data can also be rearranged and partitioned so as to minimize an entropy score that is based on the size and entropy of the data partitions. The approach is capable of enhancing performance, reducing decoding latency, and reducing power consumption.

Abstract: A method and a computer program product and an apparatus for securing communication in heterogeneous networks that include devices with different protection levels. The method comprises monitoring, by a security agent installed on a device, communication between the device and external devices. The method comprises determining a level of in-device protection for each device based on available protection thereof. The method further comprises employing, by the security agent, an associated security policy for communications originating from the device, based on the level of in-device protection; such as resources utilized for employing security policies for communications originating from devices are correlated with the protection levels thereof. The method may further comprise enabling sharing security workload between device having trusted security agents to improve performance efficiency thereof.

Abstract: Disclosed herein are memory devices, systems, and methods of encoding and decoding data. In one aspect, an encoded data chunk is received and segmented into data segments with similar features. Each segment can be decoded based on its features. Data can also be rearranged and partitioned so as to minimize an entropy score that is based on the size and entropy of the data partitions. The approach is capable of enhancing performance, reducing decoding latency, and reducing power consumption.

Abstract: A method, system and product, configured to perform: during an execution of a program, obtaining boundaries of a stack frame of a function that is currently present in a stack, wherein said obtaining the boundaries comprises: obtaining a return address of the function in the stack; determining a length of the function using a mapping of return addresses of one or more functions in the program and corresponding lengths of the one or more functions; and determining the boundaries of the stack frame of the function based on a value of a stack pointer of the stack and based on the length of the function; based on the boundaries of the stack frame of the function, determining that the stack frame is overflown; and in response to said determining that the stack frame is overflown, performing a responsive action.

Abstract: A method for generating a decision table for selecting an optimal path out of a plurality of data paths between a client and a destination server connected through a network system, each of the plurality of data paths is connected to a router configured with a unique internet protocol (IP) address is provided. The method includes for each subnet IP address of the remote destination server and each of the plurality of data paths, measuring a network proximity; factoring the network proximity measured for each of the plurality of data paths; and ranking the plurality of data paths based on a decision function computed using the factored network proximity.

Abstract: Various method and apparatus embodiments for data dependent error correction code (ECC) encoding are disclosed. In one embodiment, a data object may include multiple portions, with each portion having different characteristics. An ECC encoder may allocate error correction resources (e.g., parity bits) to the different portions at respectively different data rates (e.g., more error correction resources to some portions relative to other portions). Upon completion of the allocation, the data object and the associated error correction resources are forwarded to a storage medium for storage therein.

Abstract: A method of removing scale from a casting may include positioning the casting relative to a laser emitter. The casting may comprise a superalloy and the scale may have formed on the surfaces thereof, with the scale being a byproduct of a method of manufacturing the casting. The method may also include passing a laser beam emitted from the laser emitter across the casting such that the laser beam causes the scale to at least one of crack, break, shatter, and spall. The superalloy may be a nickel-based superalloy and the scale may include a metal carbide layer.

Abstract: Exemplary methods and apparatus are provided for implementing a deep learning accelerator (DLA) or other neural network components within the die of a non-volatile memory (NVM) apparatus using, for example, under-the-array circuit components within the die. Some aspects disclosed herein relate to configuring the under-the-array components to implement feedforward DLA operations. Other aspects relate to backpropagation operations. Still other aspects relate to using an NAND-based on-chip copy with update function to facilitate updating synaptic weights of a neural network stored on a die. Other aspects disclosed herein relate to configuring a solid state device (SSD) controller for use with the NVM. In some aspects, the SSD controller includes flash translation layer (FTL) tables configured specifically for use with neural network data stored in the NVM.